UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Subarctic nitrogen fixation in monoculture alfalfa and mixed alfalfa/grass forage swards Ball, Matthew Thomas Auric 2008

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2008_fall_ball_matthew.pdf [ 1.27MB ]
Metadata
JSON: 24-1.0066717.json
JSON-LD: 24-1.0066717-ld.json
RDF/XML (Pretty): 24-1.0066717-rdf.xml
RDF/JSON: 24-1.0066717-rdf.json
Turtle: 24-1.0066717-turtle.txt
N-Triples: 24-1.0066717-rdf-ntriples.txt
Original Record: 24-1.0066717-source.json
Full Text
24-1.0066717-fulltext.txt
Citation
24-1.0066717.ris

Full Text

   SUBARCTIC NITROGEN FIXATION IN MONOCULTURE ALFALFA AND MIXED ALFALFA/GRASS FORAGE SWARDS  by  MATTHEW THOMAS AURIC BALL  B.Sc., The University of Victoria, 2000  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  THE FACULTY OF GRADUATE STUDIES  (Soil Science)          THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   October 2008  © Matthew Thomas Auric Ball, 2008     ii Abstract   Forage growth in the subarctic is sub-optimal due to low soil nutrient levels. Forage crops in the Yukon Territory consistently require nitrogen (N) and phosphorus fertilization to meet plant requirements. Fertilization is expensive due to transportation costs and potentially harmful to the environment so alternative, more sustainable, sources of nutrients are being sought. Alfalfa is an alternative, but there is limited knowledge in the Yukon of the benefits and management of this crop as a replacement for fertilizer N. Experiments were carried out in south central Yukon during the 2005 and 2006 field seasons to examine the potential of co-inoculation of alfalfa with N-fixing Ensifer meliloti and phosphate-solubilizing Penicillium bilaii to increase the dry matter yield and N fixation of monoculture alfalfa (Medicago sativa) cv Peace and binary mixed alfalfa with smooth bromegrass (Bromus inermis) cv Carlton or timothy (Phleum pratense) cv Climax forage swards. Interactions between alfalfa inoculation and N fertilization and late season harvest treatments were assessed. The TagTeam® inoculant from Philom Bios was used as the rhizobium source which contains both Ensifer meliloti isolate NRG-34 and Penicillium bilaii isolate PB-50. Nitrogen fixation was determined using the total plant N difference method.   Alfalfa growth and nodulation was successful in the trials. Inoculation had a positive impact on N fixation, whereas urea fertilizer at 25 kg N/ha had a negative impact in most cases. In the mixed alfalfa and smooth bromegrass stand there was a positive contribution from the alfalfa in both the establishment and second year with N fixation rates of up to 14 kg/ha. In the mixed timothy and alfalfa stand the N fixation reached 35 kg/ha in the establishment year and 102 kg/ha in the second year.  In the establishment year the dry matter yield and N fixation of the TagTeam® inoculated, monoculture alfalfa plots were 3.1 t/ha and 77 kg N/ha. In the second year, the unharvested inoculated alfalfa treatment yielded 3.4 t/ha with N fixation of 66 kg/ha compared to the late harvest treatment which yielded only 1.5 t/ha and an N fixation rate of 20 kg/ha. The effects of the late season harvest are startling and reflect the importance of removing grazing animals during the fall to allow plant energy reserves to accumulate in the roots.  Fertilizer N replacement is possible with the seeding of alfalfa into existing hay stands or in monoculture.  iii Table of Contents  Abstract............................................................................................................................................. ii  Table of Contents ............................................................................................................................ iii  List of Tables ....................................................................................................................................v  List of Figures ................................................................................................................................. vii  List of Abbreviations .......................................................................................................................viii  Glossary........................................................................................................................................... ix  Acknowledgements ..........................................................................................................................x  Dedication........................................................................................................................................ xi  Co-Authorship Statement ............................................................................................................... xii  Chapter 1  Background  1.1 Introduction ...................................................................................................................1 1.2 History of Subarctic Alfalfa Research ..........................................................................4 1.3 Selecting Appropriate Alfalfa Cultivars and Mesotrophic Inoculants ............................9 1.4 Quantifying Nitrogen Fixation and Transfer in Binary Mixed Swards .........................14 1.5 Research Objectives and Hypotheses........................................................................21 1.6 References..................................................................................................................23  Chapter 2  Establishment and Management of a Subarctic Monoculture Alfalfa Sward  2.1 Introduction .................................................................................................................28 2.2 Materials and Methods................................................................................................30 2.3 Results ........................................................................................................................34 2.4 Discussion...................................................................................................................36 2.5 Conclusion ..................................................................................................................38 2.6 References..................................................................................................................39  Chapter 3  Subarctic Nitrogen Fixation in a Binary Mix Alfalfa/Smooth Bromegrass and Alfalfa/Timothy Sward  3.1 Introduction .................................................................................................................41 3.2 Materials and Methods................................................................................................42 3.3 Results ........................................................................................................................46 3.4 Discussion...................................................................................................................53 3.5 Conclusion ..................................................................................................................59 3.6 References..................................................................................................................60  Chapter 4  Conclusion  4.1 Comparison of Research Results ...............................................................................62 4.2 Strengths and Shortcomings of Research ..................................................................64 4.3 Future Research .........................................................................................................66 4.4 Concluding Statement.................................................................................................69 4.5 References..................................................................................................................70   iv   Appendix A  Experimental Conditions ...........................................................................................72  Appendix B  Detailed Methodology ................................................................................................81  Appendix C  Calculations, Conversions and Equations .................................................................88  Appendix D  Statistical Equations and Analysis of Variance Data.................................................89  Appendix E  Experimental Designs ................................................................................................94  Appendix F  Understanding Nutrient Flux using Plant-Resin Simulator Probes ............................96  Appendix G  Cooperator Farm Sites ............................................................................................102  Appendix H  Growing Alfalfa in the Yukon ...................................................................................108  Appendix I  Soil Nutrient Analyses ...............................................................................................116  References ...................................................................................................................................122  v List of Tables  Table 1.1: Nodule Ranking Criteria ................................................................................................16 Table 1.2: Comparison of Nitrogen Fixation Results from the Isotope Dilution and Nitrogen Difference Method .......................................................................................18  Table 2.1: Weather May to September Yukon Government Research Farm ................................31 Table 2.2: Modified Nodule Ranking Criteria..................................................................................33 Table 2.3: Non N-Fixing Mean Reference Value for Timothy ........................................................34 Table 2.4: Nodule Rankings by Inoculum Treatment .....................................................................34 Table 2.5: Monoculture Alfalfa Dry Matter Yields, N Content and N Fixation for 2005 and 2006..35  Table 3.1: Weather May to September at the Yukon Government Research Farm .....................43 Table 3.2: 2005 Yukon Government Research Farm Smooth Bromegrass and  Alfalfa Establishment Year Data ...........................................................................................48 Table 3.3: 2006 Yukon Government Research Farm Smooth Bromegrass and Alfalfa Establishment Year Data ...........................................................................................48 Table 3.4: 2006 Yukon Government Research Farm Smooth Bromegrass and Alfalfa Second Year Data......................................................................................................48 Table 3.5: Analysis of Variance of 2005 Dry Matter Alfalfa Yield in the Seeding Year..................49 Table 3.6: 2005 Yukon Government Research Farm Timothy and Alfalfa Establishment Year Data ...........................................................................................50 Table 3.7: 2006 Yukon Government Research Farm Timothy and Alfalfa Establishment Year Data ...........................................................................................50 Table 3.8: 2006 Yukon Government Research Farm Timothy and Alfalfa Second Year Data......................................................................................................50 Table 3.9: Mean Non N-Fixing Reference Value Smooth Bromegrass .........................................51 Table 3.10: Mean Non N-Fixing Reference Value for Timothy ......................................................51 Table 3.11: Analysis of Variance of Nitrogen Fixation Rates of Alfalfa in Timothy in Year 2.........52 Table 3.12: Smooth Bromegrass N Recovery Efficiency 2005 and 2006 ......................................52 Table 3.13: Timothy N Recovery Efficiency 2005 and 2006 ..........................................................52  Table A.1: Long Term Climate Averages from May to September in the south central Yukon .....73 Table A.2: Weather Data from May to September at the Yukon Government Research Farm.....74 Table A.3: Estimated Nutrient Removal in kg/ha per 1 tonne of Dry Matter Alfalfa for Various Regions of North America.............................................................................76 Table A.4: Volumetric Moisture Content of Fields at the Yukon Government Research Farm 2005 and 2006..................................................................................................79 Table A.5: Gravimetric vs Time Domain Reflectrometry Volumetric Moisture Content .................82 Table A.6: Modified Nodule Ranking Criteria .................................................................................85 Table A.7: Yukon Government Research Farm Smooth Bromegrass and Alfalfa Analysis of Variance Output.......................................................................................90 Table A.8: Yukon Government Research Farm Timothy and Alfalfa Analysis of Variance Output.......................................................................................91 Table A.9: Yukon Government Research Farm Monoculture Alfalfa Analysis of Variance Output.......................................................................................92 Table A.10: Rafter ‘A’ Ranch Smooth Bromegrass and Alfalfa Analysis of Variance Output ........93 Table A.11: PRS™-Probe Burial Dates and Irrigation and Moisture Levels in 2005 .....................98 Table A.12: 2005 Rafter 'A' Ranch Smooth Bromegrass (G) and Alfalfa (A) Establishment Year Data (Blocks 1-4) .....................................................................106 Table A.13: 2006 Rafter 'A' Ranch Smooth Bromegrass (G) and Alfalfa (A) Establishment Year Data (Blocks 5-7) .....................................................................106 Table A.14: 2006 Rafter 'A' Ranch Smooth Bromegrass (G) and Alfalfa (A) Second Year Data (Blocks 1-4)......................................................................................................106 Table A.15: 2005 Soil Sample Analysis from the Smooth Bromegrass and Alfalfa Sward at the Yukon Government Research Farm ..............................................................117 Table A.16: 2006 Soil Sample Analysis from the Smooth Bromegrass and Alfalfa Sward at the Yukon Government Research Farm ..............................................................118  vi  Table A.17: 2005 Soil Sample Analysis from the Timothy and Alfalfa Sward at the Yukon Government Research Farm ..............................................................119 Table A.18: 2006 Soil Sample Analysis from the Timothy and Alfalfa Sward at the Yukon Government Research Farm ..............................................................120 Table A.19: 2005 Soil Sample Analysis from the Monoculture Alfalfa Sward at the Yukon Government Research Farm ..............................................................121     vii List of Figures  Figure 1.1: Soil Temperature 0-10cm Depth in 2005 .....................................................................25 Figure 1.2: Soil Temperature 0-10cm Depth in 2006 .....................................................................26  Figure 3.1: Scatterplot of Dry Matter of Alfalfa vs Smooth Bromegrass ........................................56 Figure 3.2: Scatterplot of Dry Matter of Alfalfa vs Timothy ............................................................56 Figure 3.3: Nitrogen Contribution Comparison of Alfalfa and Grass in 2005 and 2006.................58  Figure A.1: Soil Temperature 0-10 cm Depth from May to October 2005 .....................................78 Figure A.2: Soil Temperature 0-10 cm Depth from May to October 2006 .....................................78 Figure A.3: Layout of 1) Monoculture Alfalfa ..................................................................................94 Figure A.4: Layout of 2) Smooth Bromegrass and Alfalfa Mixed Forage Stand ............................94 Figure A.5: Layout of 3) Timothy and Alfalfa Mixed Forage Stand ................................................95 Figure A.6: Layout at Rafter ‘A’ Ranch...........................................................................................95 Figure A.7: Alfalfa Growth Curve in the Smooth Bromegrass Stand 2005 ....................................98 Figure A.8: Nitrate Levels in the Soil through the 2005 Season ....................................................99 Figure A.9: Phosphate Levels in the Soil through the 2005 Season............................................100      viii List of Abbreviations  DM – Dry Matter  EC – Electrical Conductivity  ID – N15 Isotope Dilution  PRS™ – Plant Root Simulator Probes (from Western Ag Innovations Inc)  ND – Nitrogen Difference  NF – Atmospheric N2 fixation  NT – Nitrogen Transfer     ix Glossary  Cultivar – A cultivated selection that can be propagated reliably in a prescribed manner.  EGDD – Effective Growing Degree Days, the Growing Degree Days multiplied by a factor to account for the increased daylight North of 60 degrees latitude.  Evapotranspiration (ET) – The combined water loss from soil evaporation and plant transpiration.  Furrow – A long shallow trench in the ground (especially one made by a plow).  Growing Degree Days (GDD) – Can be calculated in a number of ways, the Yukon Government Agriculture Branch uses the following calculation: beginning the fifth consecutive day with mean temperatures above 5oC, and terminating the day of the first killing frost (-2.2oC) which occurs after mid-July.  Permanent Wilting Point (PWP) – The soil moisture content at which the plant will wilt and die. Any remaining water in the soil is insufficient to meet the plant requirements.  Sward – Land covered with grassy turf, such as a lawn or meadow.  Variety – A rank in botany below that of species.   x Acknowledgements  I would like to thank the following people and companies for their contribution to this research.  Dr. Art Bomke Associate Professor, Agroecology Faculty of Land and Food Systems University of British Columbia  Dr. Maja Krzic Assistant Professor, Agroecology/Forest Sciences University of British Columbia  Dr. Sue Grayston Associate Professor, Forest Sciences Department University of British Columbia  Tony Hill Director, Agriculture Branch Yukon Government  Scott Smith Land Resources Agroecology Pacific Agriculture Research Center  Dr. Stephen D. Sparrow Plant, Animal & Soil Sciences Dept. Head University of Alaska Fairbanks   Dave and Tracey Andrew Rafter ‘A’ Ranch  Bill and Barbara Drury Circle ‘D’ Ranch  Garret Gillespie and Heidi Marion Wild Blue Yonder    Field assistants   Leigh Macmillan, JR Ouellett, Melody Collins, Denis Lacroix, Sean MacDonald, Nathan Coates, and Carolyn Lewis   Funding   • Advancing Canadian Argiculture and Agri-Foods • Yukon Government • Northern Research Institute • Northern Transportation Fund   Companies   Western Ag Innovations Ltd. #3-411 Downey Road Saskatoon, SK S7N 4L8 Phone: (306) 978-0373  Philom Bios 3935 Thatcher Avenue Saskatoon, SK S7R 1A3 Phone: (306) 657-8200 Norwest Labs 7217 Roper Road Edmonton, Alberta T6B 3J4 Phone: (780) 438-5522    xi Dedication  This thesis is dedicated to the small farm.  I couldn’t have completed this project without the support of my loving parents and good friends.   xii Co-Authorship Statement  The identification and design of the research program was in conjunction with Art Bomke, Sue Grayston and Maja Krzic. Matt Ball conducted the field research, performed the analysis of data and prepared the manuscript.    1 Chapter 1 Background  1.1 Introduction  An ongoing concern in agriculture systems across Canada is the declining revenue margin on small farms. Inherent in this concern is the cost of production and, in any low nutrient soils, the cost of the application of fertilizers. Nitrogen (N) is often the growth-limiting nutrient for, and a key cost of, crop production and is always in demand in subarctic agriculture (Klebesadel 1978; Sparrow 1988). Agriculture endeavours in the Yukon Territory are no exception, consistently requiring N and phosphorus (P) fertilizers (Hill et al. 2002) to meet plant demand.  Fertilizers are imported to Yukon from neighbouring jurisdictions at significant cost, approximately doubling the cost of the unit purchase price (Andrew 2005). Purchasing and transporting fertilizer from southern Canada is usually the most expensive cost each spring for an established agricultural operation. In 2005, the cost for 1000 kg of fertilizer landed in Whitehorse, Yukon was approximately $550 CAN including shipping.  As of the spring of 2008 the cost of fertilizer had increased considerably to approximately $1000 CAN per 1000 kg. The price of fertilizer is coupled very closely to the price of fuel, both for the manufacture of the fertilizer and for its transportation. As the world moves into an oil limited economy, these costs will rise.  Conventional farms in the Yukon most often use urea formulated fertilizers to maximize shipping value. At 46% N, urea has the highest N analysis of any solid fertilizer source. Repeated use of this form of N can have negative environmental effects on soil properties and air emissions. South central Yukon, where the majority of farming occurs, has slightly alkaline soil conditions which, when combined with surface broadcast application of urea, can lead to ammonia losses through volatilization. Applying high levels of chemical fertilizers to the soil has been shown to eliminate beneficial soil organisms (Heichel and Henjum 1991; Hamel et al. 1992).  The majority of protracted environmental impacts of Yukon farming result from the processes needed to manufacture, ship, and apply chemical fertilizers. Intermediates of the N cycle which move from agricultural soils (emissions of nitrous oxides to the atmosphere) (Byrnes 1990; Topp 2003) and release of carbon compounds during transport are significant players in  2 global warming. Although Yukon farmers may not see these impacts directly, these processes are contributing to human impacts on the global environment.  By focusing on methods to reduce imported fertilizer use within the Yukon Territory, research can be proactive and mitigate future financial and environmental costs. The key to this dilemma is finding alternatives for imported fertilizers. There are a number of avenues that can be followed: examining land application of recycled nutrients, such as composts and fish wastes; looking at a systems approach with animal/plant rotations; reducing fertilizer use through best management practices; or looking at the biological N fixation from the association of bacteria with plants.  This research focused on the reduction of fertilizer N use in Yukon forage production systems through symbiotic biological N fixation by alfalfa (Medicago sativa L.) with bacteria, commonly known as rhizobia (Ensifer meliloti L. formerly Sinorhizobium) and the synergism added by the free-living, phosphate-solubilizing fungus Penicillium bilaii.  Forage production covers 1900 ha, 67% of the land in production in the Yukon, and is responsible for 50% of the gross annual revenue in the agriculture industry (Hill et al. 2002). Yukon forages are grown for the local horse population and, more recently, the growing cattle industry. The traditional Yukon grass-hay system is one-cut smooth bromegrass (Bromus inermis Leyss.) or, to a much lesser extent, timothy (Phleum pratense L.) dominant hay. Harvest occurs sometime in July often with free-range post-harvest grazing. Blended chemical fertilizers containing N, P and to a lesser extent potassium (K) as well as needed micronutrients are applied based on soil test recommendations in order to maximize yield. Fertilizer application will vary with the level of management. Under good management with optimum inputs, where fertilizer rates reach upwards of 190 kg/ha of N, grass forage yields can be in excess of 10 t/ha with 13% dry matter protein (Hill 2005). One of the key priorities for the agriculture industry is to reduce input costs for hay operations (Loeks 2003).  Alfalfa, a hardy, disease resistant, long-lived perennial species, provides a suitable alternative to  the traditional fertilized grass hay systems. Alfalfa has been grown with limited success in the Yukon, even though it is the most widely adapted agronomic legume for production agriculture in semi-arid, subarctic environments. Introduction of inoculated alfalfa into existing hay stands has the potential to fix upward of 250 kg N/ha/yr in temperate climates (Burity et al. 1989; Heichel and Henjum 1991), with rates in excess of 300 kg/ha having been recorded in Sweden, which has a similar latitude to the Yukon (Wivstad et al. 1987). Yet in  3 relatively similar environmental conditions in Alaska, Sparrow et al. (1995) found only 60 kg N/ha/yr was fixed. If N fixation rates in the range of those recorded in Alaska can be achieved in the Yukon, then N fertilizer application could be reduced by 30%.  Nutrient self-sufficiency is a goal of any agriculture cropping system. As N acquisition is one of the most important factors for plant growth, the ability to develop biological N fixation can contribute to a reduction in the dependency on imported N fertilizers. It has been proven in other areas of the northern environments that rhizobia/alfalfa symbiosis will lead to high quality and quantity forage yields (Carlsson and Huss-Danell 2003).    4 1.2 History of Subarctic Alfalfa Research  Research examining agronomic N-fixing plants in northern climates has been ongoing for over 100 years in many regions and many languages (Hanson et al. 1988). When examining research from other circumpolar regions it is important to keep in mind the uncontrollable factor of climatic variance. For example, Yukon experiences lower total precipitation and fewer heat units than any other agricultural area of Canada (Environment Canada 2008). When comparing alfalfa growth North of 60o, one can assume that any positive results from Yukon research could be duplicated in other agricultural regions, yet to assume research elsewhere can be applied to the Yukon requires analysis of the climatic conditions in which the research was carried out. There are many similar circumstances across agriculture regions North of 60o, day length and short seasons among them; however, there is a number of other factors that affect alfalfa growth to a greater degree, such as heat units and frequency of frosts, that often are dissimilar across the North of 60o latitude. There are few places with climatic conditions similar to the Yukon: the Delta Junction area in east central Alaska and across the north Pacific to Siberia are the most similar. Northern Canadian experiences in British Columbia and Alberta are also relevant in this study, especially work from the Peace River District. Research from Scandinavia and the northern mainland United States is also important and does provides comparative data, but environmental conditions are considerably different.  Research in the Yukon Territory began shortly after 1944 with the establishment of the Haines Junction Experimental Agricultural Sub Station, located at Mile 1019 of the Alaska Highway just north of Haines Junction on Pine Creek. This experimental station was overseen by the Beaverlodge Research Station, which in turn was part of a Northern Research Group that included Fort Vermillion, Alberta, Prince George, British Columbia, and Fort Simpson, Northwest Territories.  The Northern Research Group was formed to increase the efficiency of production under northern conditions and to assess the agricultural potential of areas not yet developed (Abbott et al. 1960). The site was selected because it was near the town of Haines Junction (approximately 170 km west of Whitehorse) and had access to water for irrigation. This station was mainly used to undertake research to determine crops that could be grown around the 60o latitude, including cereal and horticultural crops. The centre was also used for weather data collection, field husbandry, and soil and animal science. Studies were conducted until 1969 when the station was closed following reorganization of the federal agricultural research programs. Locating the station at Haines Junction was, in hindsight, not a good indication of Yukon conditions. Elevated by the Mount Saint Elias range, the town of Haines Junction sits in the foothills of the highest mountain in Canada and, as a result of adiabatic winds from glaciers  5 in the area, has a colder, shorter growing season than that experienced by the rest of the southern and central Yukon. Nonetheless, the research conducted at this facility provides a good indication of some of the issues that will be encountered during legume cultivation in Yukon.  Cold soils promote the growth of cold tolerant fungi and bacteria, some of which are detrimental to alfalfa growth. One of the major problems that faced the scientists at Mile 1019 was the presence of brown root rot, Phoma sclerotioides. P. sclerotioides (known as Plenodomus meliloti when the station was running) thrives in cool soils and is endemic to northern Canada. The fungus grows at subzero temperatures in the range of -7oC to 27oC (Stuteville and Erwin 1990). This fungal disease destroys the taproot at the plow pan level. Infected plants acquire an anemic appearance and are easily pulled up by hand. Symptoms include dark, necrotic lesions of the lateral and taproots, which are brown with a blackish border (McKenzie and Davidson 1975; Stuteville and Erwin 1990). Young lesions are generally brown and circular with dark perimeters. Maturing lesions generally expand asymmetrically and eventually girdle the root severing the root from the plant (Gray et al. 2003b). Rhizobium nodules can also be infected and rotted (Gray et al. 2003). The disease has been reported in Alaska, through to the Northwest Territories and south through the prairies.  Brown root rot was first reported in 1926 in Alberta by G.B. Sanford (Abbott et al. 1960). No cultivars were immune to brown root rot in early research at the Haines Junction Experimental Agricultural Sub Station. Fifty-six alfalfa cultivars were selected for breeding at the station in an attempt to develop an alfalfa resistant to brown root rot. The most promising alfalfa was a Medicago falcata variety BL1019, but it would not stay true to variety and was never made available for commercial sale (Abbott et al. 1960). M. falcata cultivars develop strong lateral rooting systems that were shown to have more persistance after the taproot has been damaged by P. sclerotioides. The germplasm of BL1019 was dominantly M. falcata cv Anik, a yellow flowered Siberian breed, that still persists at various locations in the Yukon along roadsides, and on one farm in the Haines Junction area.  Other problematic diseases that have been recorded in the Yukon include verticillium wilt and winter molds. Verticillium wilt is caused by Verticillium albo-atrum and has been identified in Sweden since the early 1900s. The disease has been in British Columbia since 1977 and has spread throughout the province. It is considered a serious disease that can reduce yields up to 50% (Stuteville and Erwin 1990). Early symptoms include V-shaped chlorosis of the leaflet tips. As the disease progresses some of the leaflets become desiccated and abscise; the infected  6 stem does not wilt and retains its green colour until all the leaves are dead. All stems of one plant may not be infected. The fungus sporulates on the base of dead stems and presents a grayish colour. Verticillium wilt is primarily found on irrigated fields (Stuteville and Erwin 1990).  There is also concern with winter molds such as Sclerotinia spp. borealis in northern soils. This fungal disease is especially damaging to seedling stands seeded in late summer. S. borealis produces a cotton weblike growth on the stems and crowns of infected plants (Stuteville and Erwin 1990). Individual stems in heavily impacted patchy areas will wilt. The fungi forms on plants during winter and spring, and cool, wet spring conditions increase the infection. As plants die the sclerotia fall to the ground and remain dormant for the summer. The spores then germinate in cool, wet fall weather and colonization continues through the winter (Stuteville and Erwin 1990). A number of options are used for control, including plowing and spring planting. Alfalfa cultivars show some resistance but not enough to control disease incidence.  Biotic issues can be controlled through developing disease resistant cultivars. M. falcata, not used in this study, was one of the earliest tolerant varieties to brown root rot. M. sativa cultivars were the least tolerant in early research (Abbott et al. 1960). Later, Peace was developed at Fort Vermilion, Alberta from the Grimm type alfalfa which is the least susceptible of the M. sativa cultivars (Stuteville and Erwin 1990).  In 1985, the Yukon Government initiated a two-year program in cooperation with a number of farmers throughout the Territory. The ‘New Crop Development Trials’ assessed eight varieties of alfalfa that were provided by the Beaverlodge research station: Anik, Peace, Beaver, Drylander and noncommercial varieties BL1019, Krasnou-6, NRG-84-4, and S7312. Overall, as was discovered 20 years earlier, the BL1019 was the best producer, followed by Anik, then Peace, then Beaver and Drylander (Bisset 1989). The inoculant used was the A’ culture from Nitragin, strain NRG-185 (Bisset 2005). The best performing varieties, BL1019 and Anik, were both M. falcata species; the others (Peace, Beaver, Drylander) were Medicago sativa.  The University of Alaska in Fairbanks, Alaska has been at the forefront of circumpolar legume research, determining if fertilizer substitution is a possibility in the far North. A considerable effort has been made in Alaska because of the awareness that one of the limiting factors in the development of economical and sustainable livestock industry in the state is the availability of a high yielding, high protein forage legume (Sparrow 1988). In past decades, progress was made by Klebesadel (Klebesadel 1978) on developing winter hardy strains of a number of legumes to survive extreme subarctic winters.  7  Studies in Alaska have shown that alfalfa, peas (Pisum sativum L.), sweetclover (Melilotus alba Lam.), red clover (Trifolium pratense L.) and white lupin (Lupinus albus L.) are all able to produce high quality forage with high yields that accumulate soil N (Sparrow et al. 1993). One of the concerns noted in the literature, regardless of the species of perennial legume, is crop winterkill (Sparrow 1988). Alfalfa N fixation rates of 20 - 44 kg/ha/yr/N in the seeding year and upwards of 60 kg/ha/yr afterwards have been measured at Delta Junction, Alaska (Panciera and Sparrow 1995; Sparrow et al. 1995).  Winter hardy Russian M. falcata Arcang. germplasm made possible the cropping of alfalfa in the northern mainland United States. M. falcata was imported into North America in the 1850s by German immigrant Wendelin Grimm. Subsequent winter hardy varieties can all be traced back to these initial imports (Barnes et al. 1988), including the cultivars grown in the Haines Junction Experimental Agricultural Sub Station trials. According to the famous Russian botanist Sinskaya,  M. falcata is a native plant of the steppe that can be traced into the northern steppe region in Siberia (Michaud et al. 1988). For over 100 years collection and modification of legumes have been ongoing in this northern Russian region. Alfalfa remains today one of the steppe's most important perennial legumes. Research in Russia continues on different traits, including high yields, fast growth and tolerance to multiple cuttings (Michaud et al. 1988).  A considerable amount of research was conducted at the Beaverlodge Research Station and the Fort Vermillion Sub Station in the Peace River District of Alberta by J. McKenzie, W. Rice, P. Olsen, and is continued today by Newton Lupwayi and Mandy Collins.  The scientists at Beaverlodge have worked with both alfalfa and rhizobia to improve performance. Rice et al. were able to isolate different rhizobia and develop new strains capable of fixing-N in colder soil temperatures, making them more useful to northern agriculture. Isolate NRG-34, used in the research that forms this thesis, is from Beaverlodge (Rice et al. 1995a; Rice et al. 1995b).  The work at Beaverlodge on cold adaptation found that cold temperatures limit areas of adaptation through sublethal or lethal winter injury. Sublethal winter injury can decrease vigor during the subsequent year with the long term effect of decreasing the stand (McKenzie et al. 1988). Alfalfa cannot survive summer air temperatures of minus 5oC nor winter air temperatures below minus 20oC (McKenzie et al. 1988). McKenzie also found that under hardening conditions, hardy cultivars going into dormancy could fix N.  8  Biological N fixation at Beaverlodge on a number of cultivars ranges from as low as 13 kg/ha in the seeding year to over 130 kg/ha in a two-cut system in a third year stand (Rice et al. 2001).  Legume cultivation has been ongoing in Scandinavian countries for centuries. Alfalfa is generally seeded as a mixture with Phleum pratense or Bromus inermis. Some breeding work of alfalfa is ongoing, including work by Wivstad in Sweden. Wivstad determined M. sativa first year production provided up to 242 kg/ha N from fixation (Wivstad et al. 1987). Anna Martensson, also in Sweden, has worked with various measurement techniques for determining biological N fixation and measurements of N transfer in mixed stands (Martensson and Ljunggren 1984a; Martensson et al. 1998).  Studies conducted in Minnesota over the past 100 years are of great importance to northern alfalfa development. The importation of winter hardy Medicago media cv Grimm by Wendelin Grimm occurred in the 1850s. The United States Department of Agriculture in Minnesota conducted the first field trials of Grimm alfalfa in North America (Barnes et al. 1988). Grimm alfalfa germplasm, a cross between M. sativa and M. falcata, is a part of the lineage of all the winter hardy strains used in the subarctic regions. In Minnesota, research on this germplasm continues today under the guidance of Gary Heichel and Carroll Vance who have shown the importance of symbiosis for derived plant N and have worked on a number of trials determining biological N fixation and transfer to grasses (Heichel and Henjum 1991; Vance et al. 1988).  The most comparable environment, that of the Delta Junction area of Alaska, has poor yields and low biological N fixation when compared to other circumpolar regions. Even with the development of cold tolerant, fall dormant, winter hardy cultivars over years of research, winterkill is often cited as a reason for crop failure in subarctic regions (Bisset 1989; Michaud et al. 1988; Sparrow 1988). This presents the challenge of not only low biological N fixation due to cool climates, but also reduced yields over time caused by winter injury.  9 1.3 Selecting Appropriate Alfalfa Cultivars and Mesotrophic Inoculants  1.3.1 Understanding Nitrogen Fixation The capabilities of legumes to fix N were first analyzed by Sergei Winogradsky, the ‘father of soil microbiology.’ Winogradsky was a Russian biologist who isolated the microorganisms which cause nitrification and in the 1880s, discovered that N-assimilating bacteria live in the root nodules of legumes (Waksman 1953).  Winogradsky found that a select group of bacteria convert inert N in air in the soil matrix into plant available forms.  Future studies determined that an enzyme, dinitrogenase, carries out the fixation of N2 into NH3. One part of the enzyme generates the reducing power by producing 8e-, called dinitrogenase reductase. The second part uses the reducing power to reduce the N2 (Dakora 2003).  Overall the net reaction is: N2 + 4 NADPH + 4H+ + 16ATP Æ 2 NH3 + H2 + 4 NADP+ + 16ADP + 16Pi  Utilizing the power of the dinitrogenase enzyme inherent to the bacteria Ensifer meliloti in combination with the legume Medicago has been common practice in many agricultural systems globally (Michaud et al. 1988; Zahran 1999).  This symbiotic relationship provides a source of N for the plant as well as many other growth promoting benefits (Dakora 2003; Matiru and Dakora 2005; Selosse et al. 2004).  Converting atmospheric N is an expensive reaction for the plant giving up 16 ATP in return for 2 NH3 ions. If there is an abundance of N in the soil solution the plant reduces the infection rate and slows N fixation. A number of other factors also reduce the N fixation rate, including cold temperatures, low pH, low moisture, and rhizobia isolate selection (Bordeleau and Prevost 1994; Zahran 1999).  Research within the last 20 years has determined that infection by E. meliloti begins with the release of flavonoids from the alfalfa, which initiate the Nod gene in the bacteria, releasing Nod factors from the bacteria that stimulate plant root hairs to curl (Vance et al. 1988). Rhizobia then set up an infection thread into the cortical cells. As the bacteria multiply they transform into bacteroids and are connected into the vascular system of the host plant (Dakora 2003).   10 For more than 50 years, scientists have been manipulating alfalfa varieties and bacterial strains to maximize the symbiotic relationship between the two in subarctic growing environments. Results of past research were examined to select the most appropriate alfalfa variety and inoculum for the cold, low nutrient, dry conditions inherent in the Yukon.  1.3.2 Alfalfa Cultivars Long term alfalfa survival has been researched in Alaska; many legumes have been seeded in long term trials, yet few have been found to be hardy enough for long term survival (Sparrow and Masiak 2004). Past experiments and information from seed growers have pointed to a number of M. sativa and M. falcata varieties that are appropriate for Yukon conditions.  Old varieties such as Ranger, Anik, and Peace (Berkenkamp et al. 1991; Bisset 2005; Sparrow et al. 1995), and a number of new varieties such as Alfagraze, Tophand, 2065 MF, and AC Nordica were all recommended (Pickseed Canada 2005).  Peace alfalfa was developed in the 1970s at the Agriculture Canada Experimental Farm, Fort Vermilion, Alberta and is a landrace cultivar predominantly from the Grimm type alfalfa. The first evaluations in the late 1970s showed that yields were up to 15% higher than Beaver (the standard in the day) (McKenzie et al. 1981). Peace is described as a persistent, hardy, long- lived variety with rapid regrowth after cutting (McKenzie et al. 1981). A further advantage of Peace is that the variety has proven to be less susceptible to Brown Root Rot (Gray et al. 2003a), a frequent problem noted in early Yukon research (Hortie et al. 1965). A disadvantage is that Peace is susceptible to bacterial wilts (McKenzie et al. 1981).  Legume demonstrations in 1991-1993 near Whitehorse singled out Peace as the best legume for feed and soil conservation properties. Peace stood up well for 3 years and established at every seeding against 7 other varieties of alfalfa: Grimm, Stampeder, Anik, Hi Pro West, Alouette, Real, and Arrow (White 1993).  In trials carried out in the Yukon in the early 1990s, Peace established well every year and produced 4.5 to 6.8 t/ha without irrigation during 3 years of research. Harvest from the second year stand yielded 4.4 t/ha with 15.8% protein and, the following year, 5.8 t/ha with 9.8% protein. When inoculated and fertilized Peace appeared to produce a reasonable amount of nodules (no data was provided), although half of the nodules formed were inactive (White 1993).  Alaska research has shown Peace to be a suitable forage crop. With yields upward of 4 t/ha, the variety has been recommended for seeding in the Fairbanks Nanene area of Alaska,  11 with the cautionary note that “long term alfalfa survivability is unreliable” (Sparrow and Masiak 2004).  Medicago falcata cv Anik was considered as a choice for these trials, but sourcing Anik seed proved to be impossible (a source of Anik was found in 2007 from a farmer in the Peace River area). Therefore M. sativa cv Peace was chosen for the trials reported herein due to its brown root rot resistance, northern hardiness and high yield in cool climates as indicated by previous experiments.  1.3.3 Mesotrophic Inoculants Medicago falcata and M. sativa form symbiotic relationships with E. medicae and E. meliloti species. The mesotrophic inoculants used in Canada are strains of E. meliloti under the group Rhizobiaceae.  The development of inoculants suitable to northern boreal areas, such as the Yukon, has dominantly been undertaken at the Agriculture and Agri-Food Canada (AAFC) Northern Research Stations (located in northern Alberta). Optimum temperature for nodule formation is from 20 to 30oC (Bordeleau and Prevost 1994), so the development of strains that are capable of nodulating in colder soil temperatures is critical to the furthering of legume acreage in northern Canada. Soil temperatures in the plow layer through the growing season North of 60o rarely reach 20oC and routinely stay below 15oC (Smith 1990).  Low temperatures delay infection and decrease nodulation and nitrogenase activity (Cralle and Heichel 1982; Rice et al. 1995a). A study by Rice and Olsen (1988) demonstrated that nodulation and N fixation of alfalfa were restricted at root temperatures below 12oC, plant growth was reduced at soil temperatures of 13oC, and effective nodule development was completely inhibited at 8oC.  A subsequent study to identify strains of E. meliloti with tolerance to cold temperatures, carried out in 1995, demonstrated that the critical temperature for development of the rhizobia/alfalfa N-fixing system is 10oC (Rice et al. 1995a).  Few bacterial strains are commercially available for use in Canada because seed companies typically only stock one strain of E. meliloti. The most common strain used as a preinoculant on alfalfa seed is E. meliloti NRG-185. E. meliloti NRG-185 has been used in past Yukon research and has been shown to nodulate and fix N in Yukon’s climate (White 1993). An important study carried out by Rice et al. (1995a) compared 226 E. meliloti isolates. Of these isolates, NRG-34 was identified as the fastest growing at 10oC, and having the most effective  12 and fewest ineffective nodules, the highest nodule fresh weight, and the third highest plant dry matter at 12oC (Rice et al. 1995a). These characteristics are of critical importance to M. sativa nodulation in cold Yukon soils.  Work by Prevost et al. (2003), investigating arctic strains that infect agronomic species, has potential, but to date success has been limited to sainfoin (Onobrychis viciifolia Scop.) and cannot be applied to alfalfa. Future research on arctic rhizobia will hopefully yield bacteria for agronomic legume inoculations. E. meliloti NRG-34 has been heralded as the most suited for northern conditions (Rice et al. 1995a), and is available as a peat-based inoculant from the Canadian inoculant company Philom Bios under the name N-Prove®. Successful nodulation of alfalfa in cold soils will require the most cold tolerant isolate, and at this time NRG-34 is the best choice.  Phosphorus levels in Yukon soils are often limited (Hill et al. 2002). To address the low P conditions, a co-inoculation with Penicillium bilaii isolate PB-50, that was developed by AAFC, was also included in the trials. PB-50 is a free-living, rhizospheric fungus that uses acetic, oxalic, lactic and glycolic acids to solubilize inorganic P sources up to 8 cm from the root (Kucey et al. 1989). In addition to increasing P in soil solution, the fungus stimulates plant growth. The mechanism of the additional growth promoting effects remains unexplained (Vessey and Heisinger 2001). Work done in temperate climates has shown PB-50 solubilizes 5 kg/ha/yr of P (Beckie et al. 1998).  Field work undertaken at the Beaverlodge Research Station in Alberta by Rice et al. (2000) on the use of NRG-34 and PB-50 demonstrated that the effects of using NRG-34 are significant on low P soils and that the addition of the PB-50 into the soil results in the highest effective nodulation and N yield. Penicillium bilaii is a naturally occuring soil fungus, but co- inoculation will stimulate greater fungal population around crop roots.  1.3.4 Benefits of Rhizobia Infection The greatest benefit of E. meliloti infection to the host plant is the N fixation in the root system, leading to an increased supply of plant available N and increased M. sativa yields and protein levels. This value of inoculation and subsequent infection has been demonstrated in multiple studies throughout the circumpolar north and the Peace Region (Martensson and Ljunggren 1984b; Rice et al. 2001; Sparrow et al. 1995; White 1993).   13 Aside from this fundamental benefit, there are a number of other important plant growth- promoting (PGP) effects of rhizobia infection. According to Dakora (2003) rhizobia increase nutrient acquisition, phytohormones (gibberellins, cytokinins, auxins), Nod factors, and riboflavin (lumichrome). Concentrations of these organic molecules increase in plants together with increased nutrient acquisition resulting in reduced stress effects on plants (Dakora 2003).  There is evidence that rhizobia also act as biocontrol agents. The accumulation of Nod factors in the rhizosphere induces the expression of genes involved in flavonoid biosynthesis and leads to an increase in phytoalexin and protection of plants against pathogens (Selosse et al. 2004). There is also evidence that rhizobia are parasitic on fungal hyphea (Dakora 2003). Overall, the greater the level of nodulation the greater the positive feedback with the increased PGP effects.   14 1.4 Quantifying Nitrogen Fixation and Transfer in Binary Mixed Swards  Biological N fixation (NF) is second only to photosynthesis as the most important biochemical process on earth (Vance et al. 1988). Prior to the introduction of industrial chemical N fertilizers, farm systems throughout the world relied on legume N contribution for plant growth, and in some systems this reliance continues today. Nitrogen acquisition is one of the most critical factors for plant production. Nitrogen contribution from NF can reduce the need for industrial N fertilizer application in most cropping systems including forages. Due to the symbiosis with N-fixing rhizobia, perennial forage legumes have great potential to increase sustainability in forage-based farming systems (Carlsson and Huss-Danell 2003) and reduce dependency on chemical N fertilizer additions.  Understanding the N contribution of legumes on farms relies on a number of methods used to quantify NF rates. There has been much research on the different measurement methods and on legume N transfer (NT) in forage swards. In a mixed forage stand, legumes provide a benefit as harvested above-ground high protein vegetation. They also provide a benefit to the below-ground NT from the root and nodule decomposition, and exudation to adjacent graminoid roots (Heichel and Henjum 1991). Nitrogen fixation and the subsequent amount of N transferred to a companion plant within a crop are important to understand in order to gauge the effect of the legume component in a mixed sward.  The concerns in the Yukon are twofold, that NF will be limited in the cool climate and that the NT will be small, making measurements of both values difficult.  The published data on the amount of N fixed per year by alfalfa vary widely. Nitrogen fixation rates in the establishment year are almost always lower than in year 2. There is also great variability in the yields in established stands because of the selection of alfalfa variety, climatic and soil conditions, and crop management (frequency of cutting, soil moisture, seeding rate, and fertilizer additions). Estimates of NF in alfalfa vary from 50 – 463 kg/ha/yr N with about 200 kg/ha/yr as the average (Vance et al. 1988). Introduction of inoculated alfalfa into an existing hay stand has the potential to fix approximately 200 kg/ha/yr in temperate climates of the northern mainland United States (Brady and Weil 2002; Heichel and Henjum 1991). Nitrogen fixation rates decrease progressively further north on the North American continent. Northern Alberta NF rates are as low as 13 kg/ha in the seeding year to over 130 kg/ha in a two-cut system in a third-year stand (Rice et al. 2001), and further north, NF rates of 60 kg/ha/yr were found in dry, sub-arctic conditions (Sparrow et al. 1995). Experiments conducted by Bell and Nutman (1971) over 30 years ago reported NF rates of 220 kg/ha in Europe, at a latitude similar  15 to the Beaverlodge Research Station in northern Alberta. According to Carlsson and Huss- Danell (2003) results from various northern latitudes indicate that comparable NF rates are as achievable around 60o as around 40o, but one must be cautious when making this statement due to the differences in climate.  1.4.1 Methods for Determining Nitrogen Fixation Nitrogen fixation rates are determined by several different methods, quite often leading to over- or under-estimation of the absolute NF. . Some of the most frequently used methods include: dry matter yield, total N difference (ND), nodule observations, as N fixation is positively correlated with these factors, and, directly, acetylene reduction assay (ARA), and N15 isotope dilution (ID) (Barnes et al. 1984).  Dry matter yield This method involves estimating the amount of NF based on the yield of harvested dry matter. For a specific season total dry matter and NF are often highly correlated (Vance et al. 1988). Work by Heichel and Henjum (1991) concluded that NF measured only from shoot biomass was significantly correlated with that measured on the whole plant (with an r = 0.99) and that this measurement underestimated the whole plant NF by on average 6% (Heichel and Henjum 1991). An equation based on dry matter for alfalfa to determine NF in temperate climates was developed by Carlsson and Huss-Danell (2003): NF = 0.021 x DM + 17 with an R2 = 0.91.  Total plant nitrogen difference method (ND) This method involves determination of NF by subtracting the total amount of plant N in the test crop from an appropriate non-nodulating reference crop (Danso 1995; Ledgard and Steele 1992). Cultivation of the non-fixing reference crop, either a non-nodulating legume or graminoid, occurs simultaneously in the same field as the N-fixing crop. These controls are most often legumes grown in absence of rhizobial inoculum or non-nodulating genotypes (Danso 1995). In other circumstances, perennial monocots are used, as is the case in the research forming this thesis. The only requirements are analysis for total N and determination of dry matter yields, making this method relatively inexpensive and easy to perform (Vance et al. 1988). The disadvantage of this test is that one has to assume similar N uptake characteristics for the fixing and non-fixing counterparts (Vance et al. 1988). This basic assumption, that the two crops absorb soil N with the same efficiency, may sometimes be inaccurate. In fact, because grasses often use soil N at higher rates than legumes, the use of the ND method can lead to  16 underestimated values of NF (Carlsson and Huss-Danell 2003; Danso 1995; Hardarson and Danso 1993; Ledgard and Steele 1992).  Nodule observations One of the most referenced methods of quantifying nodule observations was developed by Rice et al. (1977). They developed a scoring system, which includes nodule colour, numbers, position relative to the crown, and size. Nodules are cut open and examined to determine the percentage of pinkness, which indicates the presence of leghemoglobin and relative NF activity in the nodule. Position is assessed by the ratio of the number of nodules within 5 cm of the crown to the remaining nodules. The maximum score per plant is 10. These assigned numerical values for the criteria are based on previous descriptions of nodule characteristics and Rice’s experience.  Table 1.1: Nodule Ranking Criteria Characteristic Criteria Score Colour 90-100 % 4  70-89 % 3  50-69 % 2  30-49 % 1  0-29 % 0 Number 5-20/plant 3  >20/plant 2  1-5/plant 1  None 0 Position 60-100% 2 (% crown) 20-59% 1  0-19% 0 Size 3-10mm 1 (diameter) <3mm  0  >10mm 0 Rice et al. (1977)  Another measurement to add as part of the nodule assessment method is nodule weight as an indication of NF (Hardarson and Danso 1993). The nodule observation method is considered most accurate when used in parallel with another of the NF measurement methods.   17 Acetylene reduction assay (ARA) The basis for the ARA method is that the reduction of acetylene to ethylene mimics the reduction of N2 to NH3 by the nitrogenase enzyme (Vance et al. 1988). Gas chromatographic measurement of the resultant ethylene produced provides a value for NF. The test is highly sensitive, relatively simple and low cost. One shortfall of the method for field use is the large variability in the results requiring large numbers of replicates to achieve statistical significance (Vance et al. 1988).  N15 isotope dilution This method provides direct measurement of NF using N15 enriched gas to incubate and label the legume roots (Danso et al. 1993). If there is NF occurring, the level of N15 in the plants will be greater than the natural level in the atmosphere. Increased N15 translates into NF activity (Hardarson and Danso 1993). The ID method requires specialized equipment and is the most expensive field method. This method is precise and therefore requires fewer replicates than than the ARA method. (Vance et al. 1988).  Direct soil N measurements Another possible method of measurement, that this author did not find in the literature, is the use of Plant Resin Simulator (PRS) Probes (Western Ag Innovations) to monitor NT in a mixed stand. This is accomplished by maintaining PRS-probes in situ, removing the probes after a scheduled amount of time, and subsequently running analyses of the flux of N ions in the root zone over the season. Use of PRS Probes assumes that legume roots will exude a significant amount of nitrogenous material and that it will mineralize to ammonium or nitrate. It would be subject to similar limitations as NT, unless graminoids can take up significant amounts of organic N (See discussion in 1.4.4 below).  1.4.2 Comparison of the ND and ID Methods All of these methods, with the exception of the last, have been used in circumpolar legume research and are suitable for the Yukon. Comparisons of the ID and ND methods that were used by Sparrow et al. (1995) and Martensson and Ljunggren (1984b) demonstrated lower estimates of NF from the ND method when N yield in grass monocultures was used as a reference crop for N yield in legumes.      18 Table 1.2: Comparison of Nitrogen Fixation Results from the Isotope Dilution and Nitrogen Difference Methods ID (kg/ha) ND (kg/ha) ND/ID 84 79 0.94 58 52 0.9 44 34 0.77 63 43 0.68 23 15 0.65 (Martensson and Ljunggren 1984b; Sparrow et al. 1995)  According to Carlsson and Huss-Danell (2003) ND/ID values from experiments where grass monocultures were compared with legume/grass mixtures are significantly higher (P <0.001) than the ND/ID values obtained in studies where grass monocultures were compared with legume monocultures. Therefore, the NF data resulting from the ND method seems to be dependent on the stand composition: for monocultures ND can lead to underestimations, and for mixtures ND might possibly lead to overestimations of NF, as compared to ID (Carlsson and Huss-Danell 2003). Some researchers suggest that legume–grass mixtures may utilize soil N more efficiently and thus reach higher total N yields (Høgh-Jensen and Schjoerring 1997), resulting in underestimated N fixation. It is also possible that species composition changes root exudates influencing soil microbial processes (Carlsson and Huss-Danell 2003) which might possibly lead to overestimation of NF from the ND method.  A comparison of the ND and ID method by Walley (1996) showed that intercropped alfalfa fixed 121, 289 and 365 kg/ha N in the 1, 2 and 3 year of study. Mono-cropped alfalfa derived 210, 430 and 438 kg/ha N in the same study. The aboveground biomass N accounted for approximately 50% of the NF (Walley et al. 1996). Generally, the highest estimates of NT were obtained using the ID method whereas the ND method generated consistently lower estimates of NT (Walley et al. 1996).  1.4.3 Fluctuation of N Fixation over the Season There is low symbiotic activity in the first part of an establishment year as seedlings are developing and dependent on utilizing soil N (Vance et al. 1988). As legume roots expand in area and begin to initiate nodulation, the crop becomes more dependent on fixed N. One of the most important environmental factors affecting NF under field conditions is the quantity of soil N. Adding fertilizer N to legume crops decreases the amount of plant N derived from NF; but adding low levels of fertilizer N sometimes increases the total amount of N fixed (Vance et al. 1988). In  19 northern environments, with cold soils, biological rates are slower with less decomposition and slower growth, limiting NF in the spring and fall.  1.4.4 Nitrogen Transfer In low soil N environments, graminoid acquisition of soil N is important to maintain the grass longevity and quality in the sward. In a mixed stand, NT becomes one of the main sources of N for graminoids. Transfer occurs because of the cycling of soil N through the death of roots and nodules, the sloughing of those roots and nodules, and their decomposition (Burity et al. 1989; Heichel and Henjum 1991). Work by Heichel and Henjum (1991) reports that total graminoid N acquisition recorded was greater than the sum of the aforementioned activities. Therefore another possibility for N transfer was theorized; that the legume root systems likely release soil organic N that is subsequently taken up by the grass. Overall, the major routes of legume/grass NT are indirectly the release of N through the death and turnover of plant material and directly through transfer of root exudations (Burity et al 1989; Walley et al. 1996).  Estimates of NT using the ID method from alfalfa to surrounding intercropped graminoid plants vary. Depending on the composition of the sward, the ratio of grass to legume, and the soil conditions, NT may represent up to 68% of grass N content (Brophy et al. 1987; Heichel and Henjum 1991) representing 17% of the total amount of N fixed by the alfalfa component of the crop. Legume-grass mixtures become increasingly dependent upon atmospheric N with stand age. Heichel and Henjum (1991) estimated that the crop, with age, obtains up to 90% of legume N needs from the atmosphere. Research in southern climates, shows that NT is equal to the addition of 20 to 100 kg/ha N of fertilizer for the grass component, depending on proximity to the alfalfa and the growth habits of the grass (Burity et al. 1989; Mortenson et al. 2005). From studies on NT from alfalfa and birdsfoot trefoil to reed canarygrass, Brophy et al. (1987) indicated that NT occurred over a distance of 20 cm with maximum NT in areas of a high legume/grass ratio.  Nitrogen transfer in a mixed meadow bromegrass (Bromus riparius Rhem.) and alfalfa stand using the ND method with the treatment of NT in a mixed bromegrass and alfalfa stand, and a reference crop of bromegrass with non-nodulating alfalfa, showed that the N concentration in intercropped meadow bromegrass herbage was always higher than for mono-cropped meadow brome grass (Walley et al. 1996). Estimates of the proportion of total N transferred from alfalfa to meadow bromegrass ranged from 24 – 48% following the year of establishment (Walley et al. 1996).   20 Burity (1989) examined the NF and NT from alfalfa to smooth bromegrass and timothy mixed swards near Ottawa, Ontario. Nitrogen fixation ranged from 63 to 83% of plant N through 3 sequential harvests during the establishment year resulting in a total of 93 kg/ha N and increased to over 220 kg/ha N fixed by the 2nd and 3rd year. The proportion of N transferred to grass varied significantly with clipping time and ranged from 16% for the initial harvest to 49% in the final harvest in year 3. Up to 50% of the total N of the grass in these mixed swards was derived from NF. Grass N15 levels were lower in the mixtures than in pure grass stands, which indicates that N fixed by alfalfa was transferred to grass (Burity et al. 1989). Nitrogen transfer was significantly lower prior to the first harvest but increased substantially thereafter.  Decomposition rates affect the amount of N transferred from alfalfa to grass in the forage sward (Walley 1996). The decomposition rates in subarctic soils are related to a number of factors, including soil temperature and moisture. A study in Fairbanks, which has a warmer, wetter climate than the south central Yukon, indicates that the relationship between mass loss of buried forest floor materials and soil degree days was significant (r = 70 to 80%) and at 7.5 cm depth approximately 80% of the forest floor material decomposed over 2 years in a fallow field (Sparrow et al. 1992). We can expect slower rates of decomposition in the Yukon, but with irrigation a more conducive environment for decay can be created.  Transfer may not be a one-way street. Alfalfa exudates and decomposition of roots lead to grass accumulation of N, but grass decomposition and exudation can also become a nutrient source for alfalfa.          21 1.5 Research Objectives and Hypotheses  The general goal of this work is to achieve benefit by incorporating alfalfa into Yukon forage production systems. This research focused on establishing and managing the symbiotic relationship between the rhizobia Ensifer meliloti and alfalfa Medicago sativa cv Peace and the synergism added by the free-living fungi Penicillium bilaii in south central Yukon forage swards. The long term objective of this study was to examine the feasibility of legume production in the Yukon and the use of rhizobium inoculants as a replacement for N fertilizer additions.  Planting alfalfa with grasses has been common practice in the Yukon when a field is first put into production. At planting the fields are often fertilized up to 190 kg/ha N, not giving any opportunity to rhizobia/alfalfa symbiosis to initiate N fixation. On the other hand, without heavy fertilization the grass stand does not establish full cover and does not aggressively outcompete weeds for nutrients and light, which can lead to a weedy stand and sparse establishment. The approach used in the research forming this thesis was to plant the alfalfa into an existing grass stand with a no-till seeder after the grass is well established and weed populations are less of a concern.  Reducing the N level to near zero allows the alfalfa to establish amongst the grass as the grass will not have adequate soil N for expansive shoot growth and therefore reducing grass competitiveness. With a lower rate of N, the rhizobia will be encouraged to infect the root hairs because in a low soil N environment there is greater release of flavonoids from the alfalfa which initiate the infection process (Dakora 2003; Vance et al. 1988).  In order for Yukon producers to be able to incorporate alfalfa into a grass crop and realize the benefits of the legume N fixation, research must be conducted to determine the rates of N fixation we can expect in subarctic Yukon conditions and to understand the best management practices associated with the growing of these crops.  Legumes, particularly alfalfa, have long been used as forage and hay crops for livestock due to their high protein content. Alfalfa is currently shipped to the Yukon from southern agricultural areas to supply demand. This research will seek to understand alfalfa growing characteristics in the Yukon to provide best management practices for local production. Also, alfalfa and other legumes form an important component of many crop rotations. This research will help quantify the amount of fixed N which becomes available over time for subsequent crops  22 by measuring above ground biomass and observing nodule characteristics. By evolving to a reduced chemical N input system we should be able to improve soil health (Hamel et al. 1992; Heichel and Henjum 1991) and reduce energy use in forage systems (Pimentel and Pimentel 2007).  Hypotheses being tested  H1: Seeding alfalfa inoculated with Ensifer meliloti isolate NRG-34 into existing grass hay stands with no fertilizer application will result in higher forage yields and N levels over establishment with 25 kg N/ha fertilizer.  H2: Grazing and harvesting of alfalfa late in the season will lead to reduced second year growth.  H3: TagTeam® (combined Ensifer meliloti and Penicillium bilaii) inoculum on alfalfa seed will lead to N fixation that exceeds inoculation with E. meliloti alone.  H4: Incorporating inoculated alfalfa into smooth bromegrass and timothy monoculture hay stands, and limiting N fertilization to 25 kg/N ha will lead to higher N yields in the mixed hay stand compared to a monoculture grass hay fertilized at 170 kg N/ha.  In order to test these hypotheses five trials in three locations were established in May 2005. Two cooperator sites on forage production farms (Rafter A Ranch and Circle D Ranch) and the Yukon Government Research Farm (YGRF) were used, all located near Whitehorse, Yukon. All three locations are within the Takhini Valley, the main agricultural region in the Yukon. Unfortunately establishment at Circle D Ranch was very poor and the location was removed from the trial in 2005. The work at Rafter A Ranch was useful, but this site needed larger sample sizes to capture the variability exhibited in the stand so site description, methods and results from this trial are presented in Appendix G.  The investigations took place over the 2005 and 2006 growing seasons during alfalfa establishment (seeding year and the first production year).   23 1.6 References  Abbott, J. W., Hough, W. H., Tsukamoto, J. Y., Morrison, J. W. 1960. Experimental Farm Mile 1019, Alaska highway Yukon Territory Progress Report 1953-1959. Government of Canada. 64254-6: 1-20. Andrew, D. 2005. Personal Communication. Discussion of fertilizer rates and forage management. Barnes, D. K., Goplen, B. P., Baylor, J. E. 1988. Highlights in the USA and Canada. Pages 1- 24 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. First ed. ASA-CSSA-SSSA, Madison, Wisconsin. Barnes, D. K., Heichel, G. H., Vance, C. P., Ellis, W. R. 1984. A multiple-trait breeding program for improving the symbiosis for N2 fixation between Medicago sativa L and Rhizobium meliloti. Plant Soil 82(3): 303-314. Beckie, H. J., Schlechte, D., Moulin, A. P., Gleddie, S. C., Pulkinen, D. A. 1998. Response of alfalfa to inoculation with Penicillium bilaii (Provide™). Can. J. Plant Sci. 78(1): 91-102. Bell, F. and Nutman, P. S. 1971. Experiments on nitrogen fixation by nodulated lucerne. Plant Soil 35(1): 231-264. Berkenkamp, B., Bittman, S., Mccartney, D. 1991. Resistance of alfalfa cultivars to brown root rot. Can. J. Plant Sci. 71(1): 211-213. Bisset, K. 2005. Phone communication in 2005. Discussion of research conducted for the New Crop Development Trials. Bisset, K. 1989. Yukon crop development program 1989 field season report. Yukon Renewable Resources. 150 pp. Bordeleau, L. M. and Prevost, D. 1994. Nodulation and nitrogen-fixation in extreme environments. Plant Soil 161(1): 115-125. Brady, N. C. and Weil, R. R. 2002. The nature and properties of soils. Thirteenth ed. Pearson Education, Patparganji, Delhi, India. 960 pp. Brophy, L. S., Heichel, G. H., Russelle, M. P. 1987. Nitrogen transfer from forage legumes to grass in a systematic planting design. Crop Sci. 27(4): 753-758. Burity, H. A., Ta, T. C., Faris, M. A., Coulman, B. E. 1989. Estimation of nitrogen fixation and transfer from alfalfa to associated grasses in mixed swards under field conditions. Plant and Soil 114(2): 249-255. Byrnes, B.H. 1990. Environmental effects of N fertilizer use – an overview. Fert. Research 26: 209-215. Carlsson, G. and Huss-Danell, K. 2003. Nitrogen fixation in perennial forage legumes in the field. Plant Soil 253(2): 353-372. Cralle, H. T. and Heichel, G. H. 1982. Temperature and chilling sensitivity of nodule nitrogenase activity of unhardened alfalfa. Crop Sci. 22: 300-305.  24 Dakora, F. D. 2003. Defining new roles for plant and rhizobial molecules in sole and mixed plant cultures involving symbiotic legumes. New Phytol. 158(1): 39-49. Danso, S. K. A. 1995. Assessment of biological nitrogen fixation. Fertilizer Res. 42: 33-41. Danso, S. K. A., Hardarson, G., Zapata, F. 1993. Misconceptions and practical problems in the use of N15 soil enrichment techniques for estimating N2 fixation. Plant Soil 152: 25-52.  Environment Canada. 2008. Canadian climate normals 1971-2000. www.weatheroffice.gc.ca. Gray, F. A., Hollingsworth, C. R., Koch, D., Heald, T. 2003. Brown root rot of alfalfa. Plant Sciences Timely Information Series 1: 1-5. Hamel, C., Furlan, V., Smith, D. L. 1992. Mycorrhizal effects on interspecific plant competition and nitrogen transfer in legume-grass mixtures. Crop Sci. 32: 991-996. Hanson, A. A., Barnes, D. K., Hill, J. R. R., eds. 1988. Alfalfa and alfalfa improvement. American Society of Agronomy, Inc; Crop Science Society of America, Inc; Soil Science Society of America, Inc, Madison, Wisconsin, USA. Hardarson, G. and Danso, S. K. A. 1993. Methods for measuring biological nitrogen fixation in grain legumes. Plant Soil 152(1): 19-23. Heichel, G. H. and Henjum, K. I. 1991. Dinitrogen fixation, nitrogen transfer, and productivity of forage legume-grass communities. Crop Sci. 31: 202-208. Hill, T., Beckman, D., Ball, M., Smith, P., Whelan, V. 2002. Yukon agriculture state of the industry, 2000-2001. Yukon Government. 56 pp. Hill, T. 2006. Personal Communication. Discussion regarding forage management, Yukon soil conditions and fertilizer use. Høgh-Jensen, H. and Schjoerring, J. K. 1997. Interactions between white clover and ryegrass under contrasting nitrogen availability: N2 fixation, N fertilizer recovery, N transfer and water use efficiency. Plant Soil 197(2): 187-199. Hortie, H. J., Tsukamoto, J. Y., Gubbels, G. H. 1965. Research report 1960-1964 experimental farm. Canada Department of Agriculture. A56-432: 1-12. Klebesadel, L. J. 1978. Biological nitrogen fixation in natural and agricultural situations in Alaska. Agroborealis January: 9-12. Kucey, R. M. N., Janzen, H. H., Leggett, M. E. 1989. Microbially mediated increases in plant- available phosphorus. Adv. Agron. 42: 199-228. Ledgard, S. F. and Steele, K. W. 1992. Biological nitrogen-fixation in mixed legume grass pastures. Plant Soil 141(1-2): 137-153. Loeks, D. 2003. Strategic analysis of the Yukon agriculture industry. Yukon Government. pp 48. Martensson, A. M. and Ljunggren, H. D. 1984a. A comparison between the acetylene reduction method, the isotope dilution method and the total nitrogen difference method for measuring nitrogen fixation in lucerne (Medicago sativa L.). Plant Soil 81(2): 177-184.  25 Martensson, A. M. and Ljunggren, H. D. 1984b. Nitrogen fixation in an establishing alfalfa (Medicago sativa L.) ley in Sweden, estimated by three different methods. Applied Environ. Microb. 48(4): 702-707. Martensson, A. M., Rydberg, I., Vestberg, M. 1998. Potential to improve transfer of N in intercropped systems by optimising host-endophyte combinations. Plant Soil 205(1): 57-66. Matiru, V. N. and Dakora, F. D. 2005. Xylem transport and shoot accumulation of lumichrome, a newly recognized rhizobial signal, alters root respiration, stomatal conductance, leaf transpiration and photosynthetic rates in legumes and cereals. New Phytol. 165(3): 847- 855. McKenzie, J. S. and Davidson, J. G. N. 1975. Prevalence of alfalfa crown and root diseases in the Peace River region of Alberta and British Columbia. Canadian Plant Disease Survey 55: 121-125. McKenzie, J. S., Paquin, R., Duke, S. H. 1988. Cold and heat tolerance. Pages 259-302 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. First ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA. McKenzie, J. S., Pankiw, P., Siemens, B. 1981. Peace alfalfa. Can. J. Plant Sci. 61: 473-474. Michaud, R., Lehman, W. F., Rumbaugh, M. D. 1988. World distribution and historical development. Pages 25-91 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. First ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA. Mortenson, M. C., Schuman, G. E., Ingram, L. J., Nayigihugu, V., Hess, B. W. 2005. Forage production and quality of a mixed-grass rangeland interseeded with Medicago sativa ssp falcata. Range Ecology Manage. 58(5): 505-513. Panciera, M. T. and Sparrow, S. D. 1995. Effects of nitrogen fertilizer on dry matter and nitrogen yields of herbaceous legumes in interior Alaska. Can. J. Plant Sci. 75: 129-134. Pickseed Canada. 2005. Personnal Communication. Discussion of appropriate varieties for the Yukon. Pimentel, D. and Pimentel, M. 2007. Food, energy, and society. 3rd ed. CRC Press, New York. 380 pp. Prevost, D., Drouin, P., Laberge, S., Bertrand, A., Cloutier, J., Levesque, G. 2003. Cold- adapted rhizobia for nitrogen fixation in temperate regions. Can. J. Bot. 81(12): 1153-1161. Rice, W. A. and Olsen, P. E. 1988. Root-temperature effects on competition for nodule occupancy between 2 Rhizobium meliloti strains. Biol. Fertility Soils 6(2): 137-140. Rice, W. A., Olsen, P. E., Collins, M. M. 1995a. Symbiotic effectiveness of Rhizobium meliloti at low root temperatures. Plant Soil 170: 351-358. Rice, W. A., Olsen, P. E., Leggett, M. E. 1995b. Coculture of Rhizobium meliloti and a phosphorus-solubilizing fungus (Penicillium bilaii) in sterile peat. Soil Biology & Biochemistry 27(4-5): 703-705. Rice, W. A., Penney, D. C., Nyborg, M. 1977. Effects of soil acidity on rhizobia numbers, nodulation and nitrogen-fixation by alfalfa and red-clover. Can. J. Soil Sci. 57(2): 197-203.  26 Rice, W. A., Olsen, P. E., Lupwayi, N. Z., Clayton, G. W. 2001. Field comparison of pre- inoculated alfalfa seed and traditional seed inoculation with inoculant prepared in sterile or non-sterile peat. Commun. Soil Sci. Plant Anal. 32(13-14): 2091-2107. Rice, W. A., Lupwayi, N. Z., Olsen, P. E., Schlechte, D., Gleddie, S. C. 2000. Field evaluation of dual inoculation of alfalfa with Sinorhizobium meliloti and Penicillium bilaii. Can. J. Plant Sci. 80: 303-308. Selosse, M. A., Baudoin, E., Vandenkoornhuyse, P. 2004. Symbiotic microorganisms, a key for ecological success and protection of plants. Comptes Rendus Biologies 327(7): 639-648. Smith, C. A. S. 1990. Nature of the cryic thermal regime of agricultural soils in the Yukon Territory, Canada. In Proceedings International Symposium on Frozen Soil Impacts on Agricultural, Range and Forest Lands 90-1: 11-20. Sparrow, S. D. 1988. Inoculation of alfalfa in Alaska. Agroborealis 20(1): 38-40. Sparrow, S. D. and Masiak, D. T. 2004. Forage crop variety trials in the Tanana Valley of interior Alaska. AFES Circular 125: 1-32. Sparrow, S. D., Cochran, V. L., Sparrow, E. B. 1995. Dinitrogen fixation by seven legume crops in Alaska. Agron. J. 87: 34-41. Sparrow, S. D., Cochran, V. L., Sparrow, E. B. 1993. Herbage yield and nitrogen accumulation by seven legume crops on acid and neutral soils in a subarctic environment. Can. J. Plant Sci. 73: 1037-1045. Sparrow, S. D., Sparrow, E. B., Cochran, V. L. 1992. Decomposition in forest and fallow subarctic soils. Biol. Fertil. Soils 14: 253-259. Stuteville, D. L. and Erwin, D. C., eds. 1990. Compendium of alfalfa diseases second edition. The American Phytopathological Society. 84 pp. Topp, E. 2003. Bacteria in agricultural soils: Diversity, role and future perspectives. Can. J. Soil Sci. 83(3): 303-309. Vance, C. P., Heichel, G. H., Phillips, D. A. 1988. Nodulation and symbiotic dinitrogen fixation. Pages 229-258 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. First ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA. Vessey, J. K. and Heisinger, K. G. 2001. Effect of Penicillium bilaii inoculation and phosphorus fertilisation on root and shoot parameters of field-grown pea. Can. J. Plant Sci. 81(3): 361- 366. Waksman, S. A. 1953. Sergei Nikolaevitch Winogradsky: 1856-1953. Science 118(3054): 36- 37. Walley, F. L., Tomm, G. O., Matus, A., Slinkard, A. E., van Kessel, C. 1996. Allocation and cycling of nitrogen in an alfalfa-bromegrass sward. Agron. J. 88: 834-843. White, M. 1993. Yukon agricultural research and demonstration report 1992/93. Yukon Renewable Resources. PR 94-3: 1-10.  27 Wivstad, M., Martensson, A. M., Ljunggren, H. D. 1987. Field measurement of symbiotic nitrogen fixation in an established lucerne ley using N15 and an acetylene reduction method. Plant Soil 97(1): 93-104. Zahran, H. H. 1999. Rhizobium-legume symbiosis and nitrogen fixation under severe conditions and in an arid climate. Microbiology and Molecular Biology Reviews 63(4): 968-989.  28 Chapter 2 Establishment and Management of a Subarctic Monoculture Alfalfa Sward 1   2.1 Introduction  Nitrogen is often the growth-limiting nutrient for optimum crop yields and is consistently low in subarctic soils (Klebesadel 1978; Sparrow 1988), with background nitrate levels below 4 ppm in most south central Yukon soils (Hill 2006). Nitrogen fertilizer is expensive in the North with the additional cost of transportation above the purchase price. Urea-based fertilizers are imported from southern jurisdictions; in spring 2008, the cost of 1000 kg of fertilizer landed in Whitehorse, Yukon was approximately $1,100 CAN.  Yukon forages are supplied to the local horse and, more recently, the growing cattle industry. As is the case in Alaska (Sparrow 1988), one of the limits to economic beef production in Yukon is the availability of high yielding, high protein forage legumes capable of cold climate dryland production for hay and/or grazing. The traditional Yukon grass-hay system is one-cut smooth bromegrass (Bromus inermis Leyss.) or, less often, timothy (Phleum pratense L.) dominant hay. Harvest occurs sometime in July often with free-range post-harvest grazing. Under optimum conditions, with adequate irrigation and N fertilizer rates reaching upwards of 200 kg/ha, the resulting grass hay yields are in excess of 10 t/ha. Unfortunately the cost of producing this crop is high, and will only continue to increase with the rise of fertilizer, transportation and diesel costs.  A number of local producers have attempted to grow alfalfa (Medicago sativa L.) to produce high protein forages and reduce fertilizer N requirements. Unfortunately, there have been few positive results. It is often stated that, in monoculture, alfalfa production drops off after two years (Andrew 2005). Although alfalfa is a hardy, long- lived perennial species, the Yukon’s dry and cold climatic conditions are on the edge of this species’ tolerances. Alfalfa fields in the Yukon are generally seeded without onsite inoculation and often grazed late in the fall before soil freeze-up.  1 A version of this chapter will be submitted for publication. Ball, M.T.A. and Bomke, A.A. Establishment and Management of a Subarctic Monoculture Alfalfa Sward. 29  In order to establish the alfalfa/rhizobium symbiosis plant roots must come into contact with, and be infected by, the appropriate rhizobium. Treating seed onsite with peat-based inoculum ensures high levels of rhizobium on each seed, positioning the rhizobium to infect the root of the alfalfa after plant germination. Pretreated seed is the most commonly used method of inoculation today in the Yukon (Vanderkley 2005). Rice et al. (2001) compared onsite peat-based inoculation with preinoculated seed and found that rhizobium numbers are usually greater on freshly inoculated seeds than on pre- inoculated seeds, leading to the assumption that onsite seed inoculation provides greater assurance that the rhizosphere will contain adequate rhizobia for rapid and substantial infection.  There are a number of different E. meliloti strains used to inoculate alfalfa. Most frequently used in the Yukon is pre-inoculated seed with Nitragin cultures. For previous Yukon research the inoculant used was Nitragin A’ culture, strain NRG-185. This strain is still common on pretreated seed imported from Alberta or British Columbia. Peat based TagTeam® from Philom Bios, containing E. Meliloti NRG-34, offers an alternate choice, and was used in this research. This inoculum includes Penicillium bilaii PB-50 to gain the extra benefit of the fungal inoculant for soil P mineralization. Field work undertaken at the Beaverlodge Research Station by Rice et al. (2000) on the use of NRG-34 and PB-50 demonstrated the effectiveness of using NRG-34 and that the addition of the PB-50 into the inoculum results in the highest effective nodulation and N yield. Penicillium bilaii is a naturally occurring soil fungus, but co-inoculation can build a greater fungal population around crop roots.  In an establishment year for a perennial hay crop, producers look to benefit from the new sward. Harvesting the first year crop early in the season would have little benefit as perennial crops in the Yukon. Since crops tend to require a full season before biomass production is substantial enough for harvest, producers will often graze animals late in the first season.  The majority of late season grazing in the Yukon is by horses. These selective grazers will eat alfalfa and other legumes preferentially to other plants in an open grazing system. Removing the leaves from alfalfa in the late fall reduces the 30 synthesizing and translocation of substances to the crowns and roots for overwintering energy reserves (McKenzie et al. 1988). Agricultural extension agents in southern Canada often recommend harvesting alfalfa 4 – 6 weeks before the first killing frost. Cutting close to fall dormancy interferes with the accumulation of food reserves because new growth is produced at the expense of winter reserves. Winterkill is the term used to describe the failure of an overwintering crop to survive low temperatures and concomitant cold injury (Leep et al. 2001). A plant's capability to overwinter is based on the ability to go dormant, and to survive both abiotic and biotic stresses through the winter including frost, frost heave, ice formation, low temperatures, and fungal and bacterial infection (Bertrand and Castonguay 2003). Those cultivars that have a greater winter hardiness rating tend to persist longer in the stand, but it is expected that none would last through multiple late cuttings.  The main objective of this study was to evaluate the effect of inoculation and late season grazing in monoculture alfalfa establishment. The overall goal of this study was to examine Dry matter yield, N content and N fixation of monoculture alfalfa cv Peace to determine the feasibility of alfalfa production, with associated N fixation, in the south central Yukon.  2.2 Materials and Methods  This study was conducted during the 2005 and 2006 field seasons at the Yukon Government Research Farm, near Whitehorse, Yukon. Yukon climatic conditions are conducive to cool season crops. Growing season frosts are frequent, the research site is located within a small cleared area that causes increased frost occurrences, with the frost free period ranging from 25 – 33 days during the two years of study (Table 2.1). The growing degree days at the research site during the two years of the experiment averaged 722. The Yukon is classified as a semi-arid environment with precipitation averaging 300 mm/yr, part of which falls as snow in the winter and is lost through spring runoff and sublimation (Environment Canada 2008).     31  Table 2.1: Weather May to September Yukon Government Research Farm  Precipitation (mm) Irrigation (mm) ET (mm)1 GDD2 Mean Air Temp (oC) Frost Free (Days) 2005 171 108 405 659 12.2 25 2006 134 80 344 785 10.6 33 Historical Normals 160 * --- --- 743 ** 10.5 * 36 ** 1. Evapotranspiration is the actual ET determined from an evapotranspiration simulator (Etgage Company) 2. The number of growing degree days (GDD) are calculated beginning the fifth consecutive day of the year with daily mean temperatures above 5 C, and terminated the day of the first killing frost (-2.2 C) occurring after July 15. This killing frost temperature does not need to occur as a daily mean temperature, but rather at any moment of a day. * Historical Normals calculated from Environment Canada 30 Year Normal data ** Historical Normals calculated from long term climate monitoring at the Yukon Government Research Farm  Weather during the years of research was near to climate normals (Table 2.1). The 2005 growing season began with a warmer than normal spring (April and May) leading to an early start to the growing season. Very little precipitation fell in May and the daily temperatures were above normal. The temperatures remained above average until mid June at which time temperatures dropped to seasonal averages. Precipitation was below the seasonal normal until the end of June when a record breaking one day rainfall of 27 mm brought the precipitation total for the month above normal at 62 mm. July and August temperatures and rainfall returned to above normal values. The following year, 2006, was much cooler at the beginning of the season, resulting in slower spring growth. Although the monthly temperature in May was still above normal, it was cooler than the previous spring. In 2006 mean air temperatures were below normal for the overall season due to well below normal temperatures in August.  The soils at the site are Orthic Brunisols, developed from fine-textured glacio- fluvial deposits underlain by deep glacio-lacustrine deposits, and are classifed as loam, with an average particle size breakdown of 42% sand, 47% silt and 11% clay. The average soil pH is 7.0 and has been trending downwards over time with agricultural activity (Hill et al. 2002). Average soil organic matter is around 2% by weight, typical of the Takhini Valley. The soil is well drained and stone free. These soils belong to the Lewes soil association and are described as having 7.5 to 15 cm of brown loam over 7.5 32 to 10 cm of yellowish brown non-calcareous loam over calcareous bedded silty clay with good moisture holding capacity (Day 1962; Smith 1990).  The site was cleared in 1987 of willow (Salix spp), aspen (Populus tremuloides Michx.), spruce (Picea spp), lodgepole pine (Pinus contorta Douglas ex Louden), soap berry (Shepherdia spp), and bearberry (Arctostaphylos spp). Since being cleared, the site has been used intensively for a variety of research projects. The experimental plots are located on a level field surrounded by a dominantly lodgepole pine forest, which provides shelter from winds, but consequently creates greater frost occurrences and a slightly shorter season than is experienced in other large cleared fields of the south central Yukon. Aside from the frost occurrence, the soil, landscape and climatic properties of the site are typical of those encountered at many farms in the south central region of the Yukon. The elevation of this site is 660 m above mean sea level.  Seed was handsown in 20 cm spacing at 15 kg/ha in a completely randomized split plot design in eight 2x2 m plots with 1 m spacing between plots. Main plot treatments were a TagTeam® inoculant from Philom Bios Inc. (Saskatoon, Saskatchewan) (containing NRG-34 Ensifer meliloti and PB-50 Penicillium bilaii) and a control with no inoculation. A split plot treatment was applied to each experimental unit at the end of the first season to determine the effect of late season grazing in the establishment year. Half of each plot was clipped in late August 2005 to mimic late season grazing that is often associated with substantial legume losses in subsequent year’s yields. In the other half of each experimental unit the shoot biomass was clipped after the soil was frozen.  Fertilizers were surface broadcast 5 days after seeding to meet soil test recommendations. Potassium chloride (0-0-62) was applied at a rate of 35 kg/ha and boron (as Borax) was applied at a rate of 1 kg/ha in the spring of year 1. Although N was also lacking, no N was added in order to determine N fixation rates. The crop was irrigated weekly to meet some of the crop water demand (108 mm was applied in 2005 and 80 mm in 2006).  Alfalfa cv Peace was chosen for these trials due to its brown root rot resistance, northern/winter hardiness, high yield and previous track record with Yukon producers. 33  Detailed nodule assessments were carried out on September 19, 2005 and October 6, 2006. Four plants with roots were randomly extracted with a shovel from each experimental unit. Soil and plants were soaked in water to loosen the soil from the roots. Each plant was assessed using a modified ranking system developed by Rice et al. (1977). The ranking system included scores for nodule numbers, position relative to the crown, and size. Nodules were cut open and examined to determine the degree of pinkness which indicates the presence of leghaemoglobin. The colour value was difficult to assess because of the time of year, therefore colour was not used in the analysis.  Nodule position was assessed by the ratio of the number of nodules within 5 cm of the crown to the remaining nodules. The maximum score without the colour being quantified was six (Table 2.2).  Table 2.2: Modified Nodule Ranking Criteria Characteristic Criteria Score Number 5-20/plant 3  >20/plant 2  1-5/plant 1  None 0 Position 60-100% 2 (% crown) 20-59% 1  0-19% 0 Size 3-10mm 1 (diameter) <3mm  0  >10mm 0 (Rice et al. 1977)  Above ground biomass of alfalfa was harvested from randomly selected rows at 5 cm height to determine the dry matter weight, N content, and N fixation. Samples were clipped from selected 2 m rows, sorted into stage of growth to determine percent bloom, dried for 48 hours at 40oC, and weighed to determine “as fed weight.” One hundred gram composite subsamples were sent to Norwest Labs in Lethbridge, Alberta for dry matter determination and LECO protein analysis. Harvest occurred on August 28, 2005 at 10% bloom and July 31, 2006 at 40% bloom. 34  Nitrogen content was calculated from the percent N in the composite plant subsample multiplied by the DM weight.  Nitrogen fixation was calculated using the total plant N difference method (ND). This method requires a non-nodulating reference crop be grown. Unfortunately the uninoculated control plots became nodulated in the first year potentially through native rhizobium, but more likely through the movement of NRG-34 in soil water and dust into the control plots. A timothy reference crop with no fertilizer was used in this experiment instead (Table 2.3).  Table 2.3: Non N-Fixing Mean Reference Value for Timothy  2005 2006 Mean (kg/ha/yr) 18.0 23.0 SD (kg/ha/yr) 4.7 4.6 CV% 26.1 19.9  All data were compared using an analysis of variance model in a completely randomized design in year 1 and a split plot completely randomized design in year 2 using SAS® 9.1 (SAS Institute). Treatment means were compared using P = 0.05 and the conservative Bonferroni multiple comparisons method.  Bonferroni testing is a stepwise, sequentially rejective, method for hypothesis testing.  2.3 Results  Detailed nodule assessments were carried out to determine the effects of inoculation on alfalfa root nodule numbers in the seeding year (2005) and second year (2006).  Table 2.4: Nodule Rankings by Inoculum Treatment  TagTeam® Control Year 1 Sept 2005 not cut 5.5 a 3.3 b Year 2 Oct 2006 not cut 4.7 ab 4.5 ab 35 * Different suffix letters refer to numbers that are statistically significant at P = 0.05 based on Bonferroni critical distances.  The inoculated plots scored substantially higher than the control plots in the initial assessment year, scoring between 5.25 – 6 points with a mean of 5.5 compared to a mean of 3.3 for the uninoculated plots (Table 2.4). In the second year, the scores between treatments were very similar. This is not surprising since both inoculated and control plots showed nodulation by mid season 2005. Nodulation was typically evident within one month of seeding. Nodulation in the control plots was unlikely to be from native rhizobium populations as the nodulation appeared to coincide with a rain event that occurred 2 weeks previous to the first evidence of nodulation on the controls. Soil water movement from this event or wind may have contributed to the contamination of the controls from the neighbouring inoculated treatments.  Table 2.5: Monoculture Alfalfa Dry Matter Yields, N Content and N Fixation during 2005 and 2006 Alfalfa Establishment Year (2005)  DM Yield N Content NF  t/ha/yr kg/ha/yr kg/ha/yr TagTeam®  3.1 a  95 a  77 a Control  2.5 a  83 a  65 a Alfalfa 2nd Year (2006)  DM Yield N Content NF  t/ha/yr kg/ha/yr kg/ha/yr TagTeam® Cut  1.6 b  43 b  20 b TagTeam® Not Cut  3.4 a  89 a  66 a Control Cut  1.5 b  43 b  20 b Control Not Cut  2.8 ab  73 ab  50 ab * Each value is the mean of 4 samples. The lettering is to be examined by column within each year. Values with the same letter are not significantly different based on the Bonferroni analysis at P = 0.05.  The DM and N yields of the TagTeam® inoculated plots in the establishment year were 3.1 t/ha and 95 kg/ha of N (Table 2.5). The uninoculated plots, that eventually became inoculated mid-season 2005 provided 2.5 t/ha of DM and 83 kg/ha of N. In year two the yields of the uncut treatments were 3.4 and 2.8 t/ha with yields of only 1.5 to 1.6 36 t/ha in the late season harvest treatment. The split plot late season harvest treatment was significant in the ANOVA and, based on the Bonferroni analysis, in the TagTeam® split plots as well.  The TagTeam® inoculated plots that were not harvested in August 2005 had DM of 3.4 t/ha yielding 89 kg N/ha N in year two, whereas the harvested plots yielded DM 1.6 t/ha and only 43 kg N/ha.  2.4 Discussion  The inoculation of alfalfa with the Philom Bios TagTeam® product led to high nodulation scores, and total DM yield of over 3 t/ha/yr and NF of 65-75 kg/ha/yr in the establishment year. The DM yields achieved in the inoculated unharvested treatments are comparable to a first cut in a two cut system from Beaverlodge (Rice et al. 2000), and the N content is slightly higher than single cut yields observed at Delta Junction in Alaska in which 60 kg/ha/yr of N were found (Sparrow et al. 1995). The NF rates are slightly higher than rates achieved in the Delta Junction area, which were 23-44 kg/ha/yr, but are similar to the rates achieved in the Fairbanks area using the ID method (Sparrow et al. 1995). Results from other northern regions have shown similar NF rates to Alaska in the establishment year. These rates are at the low end of the range from work in southern Canada and into the northern US with N fixation in the seeding year between 62 to 93 kg/ha/yr (Burity et al. 1989; Heichel et al. 1981; Heichel and Henjum 1991; Walley et al. 1996).  Work by Walley et al. (1996) and Burity et al. (1989) into second year growth of alfalfa showed a doubling in the NF rates compared to the establishment year, increases that were not observed in this research. In year two of the study NF rates ranged from 50 to 65 kg/ha, slightly lower than in establishment year. This is likely due to the loss of nodules over winter, shown from the reduced nodule scores in year two in the TagTeam® treatment (Table 3.4). One of the overwintering issues noted in research from Beaverlodge, Alberta, includes a decline in nodule occupancy from fall to the following summer (Rice et al. 2001) resulting in reduced NF the subsequent year. Other reasons that increased NF did not occur into year two may include the cooler growing season in 2006, or lower soil water content during June and July in 2006 (Table 2.4). 37  The ability to compare a control non-nodulated alfalfa stand was compromised midway through the first season, due to either a healthy native rhizobia population or, more likely, the movement of NRG-34 rhizobia in the soil from the inoculated stand in surface and subsurface water flow or wind. It is possible that there was a large native rhizobia population, which could have impacted nodule formation and NF rates. No testing was available to confirm the bacterial DNA. In environments with high native rhizobia populations an imported inoculum, such as NRG-34, could have less of an impact on nodule growth (Rice et al. 2000).  In the absence of a non-nodulating reference crop the NF values were calculated using timothy. This may have led to an underestimation of the N fixation in the alfalfa stand as the rate of NF is assumed to equal the difference between N accumulated in the fixing crop and the reference crop. The disadvantage of a timothy reference crop is the assumption of similar N uptake for the fixing and non-fixing crops (Vance et al. 1988). This basic assumption, that the two crops absorb soil N with the same efficiency, may sometimes be inaccurate because grasses often use soil N at higher rates than legumes (Danso 1995). The use of the ND method can lead to underestimation of NF (Carlsson and Huss-Danell 2003; Hardarson and Danso 1993; Ledgard and Steele 1992). Nonetheless, in this research, using a timothy reference crop resulted in NF rates that are comparable to northern research from Alaska by Sparrow et al. (1995).  To mimic the late season grazing treatment, the plots were clipped at the end of August in 2005. Most Yukon hay crops are not cut in the first year, but rather grazed late in the season, often before freeze up. Emulating late fall grazing, as was done in this research, resulted in greatly reduced yields in the subsequent year’s growth. The effects of the late season harvest are startling and reflect the importance of removing grazing animals during the fall period before dormancy. Dry matter yields of alfalfa for the late season harvest treatment were half of the unharvested yields in the second year and NF values were down to a third of the fixation achieved by the unharvested samples. Research by Panciera (1998) on late season harvesting observed decreased energy reserves in plants harvested late in the season compared with earlier harvests or unharvested plants.  38 The correlation of N fixation (kg/ha/yr) with DM yield (kg/ha/yr) was calculated to determine the ability to predict N fixation based on the total dry matter harvested. The resultant R2 of 0.991 for the second year harvest shows a very good fit for the predictive model. In this monoculture crop, the equation NF = 0.0254xDM – 20.058 provides an accurate calculation of NF based on DM. For a specific season total DM and NF are often highly correlated (Vance et al. 1988). An equation based on DM to determine NF in northern Sweden by Carlsson and Huss-Danell (2003) was NF = 0.021xDM + 17 for alfalfa. The equation derived from our study is similar, but crosses the Y-axis at a negative value. More study in this climate will be required to determine the accuracy of this equation over time.  2.5 Conclusions  Late season grazing of alfalfa in an establishment year allows for an economic use of the crop in the first year. Unfortunately, the result of this late season activity, as highlighted by this experiment and discussed in the literature, removes the leaves that would otherwise be synthesizing substances that could be translocated to the crowns and roots for energy reserves. This is a critical period for the accumulation of food reserves because when the plant is cut, new growth is produced at the expense of winter reserves (McKenzie et al. 1988). This reduces plant hardiness, and increases the likelihood of winterkill, leading to reduced yields in subsequent years.  39 2.6 References  Andrew, D. 2005. Personal Communication. Discussion of fertilizer rates and forage management. Burity, H. A., Ta, T. C., Faris, M. A., Coulman, B. E. 1989. Estimation of nitrogen fixation and transfer from alfalfa to associated grasses in mixed swards under field conditions. Plant Soil 114(2): 249-255. Carlsson, G. and Huss-Danell, K. 2003. Nitrogen fixation in perennial forage legumes in the field. Plant Soil 253(2): 353-372. Danso, S. K. A. 1995. Assessment of biological nitrogen fixation. Fert. Research 42: 33- 41. Day, J. H. 1962. Reconnaissance soil survey of the Takhini and Dezadeash Valleys in the Yukon Territory. 537: 14-2. Environment Canada. 2008. Canadian climate normals 1971-2000. www.weatheroffice.gc.ca. Hardarson, G. and Danso, S. K. A. 1993. Methods for measuring biological nitrogen fixation in grain legumes. Plant Soil 152(1): 19-23. Heichel, G. H. and Henjum, K. I. 1991. Dinitrogen fixation, nitrogen transfer, and productivity of forage legume-grass communities. Crop Sci. 31: 202-208. Heichel, G. H., Barnes, D. K., Vance, C. P. 1981. Nitrogen fixation of alfalfa in the seeding year. Crop Sci. 21: 330-335. Hill, T. 2006. Personal Communication. Discussion regarding forage management, Yukon soil conditions and fertilizer use. Klebesadel, L. J. 1978. Biological nitrogen fixation in natural and agricultural situations in Alaska. Agroborealis January: 9-12. Ledgard, S. F. and Steele, K. W. 1992. Biological nitrogen-fixation in mixed legume grass pastures. Plant Soil 141(1-2): 137-153. McKenzie, J. S., Paquin, R., Duke, S. H. 1988. Cold and heat tolerance. Pages 259- 302 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. 1st ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA. Panciera, M.T. 1998. Factors affecting cold hardiness development. Alaska Cooperative Extension Publication FGV-00143. Fairbanks Alaska. 12 pp. Rice, W. A., Penney, D. C., Nyborg, M. 1977. Effects of soil acidity on rhizobia numbers, nodulation and nitrogen-fixation by alfalfa and red-clover. Can. J. Soil Sci. 57(2): 197-203. 40 Rice, W. A., Lupwayi, N. Z., Olsen, P. E., Schlechte, D., Gleddie, S. C. 2000. Field evaluation of dual inoculation of alfalfa with Sinorhizobium meliloti and Penicillium bilaii. Can. J. Plant Sci. 80: 303-308. Rice, W. A., Olsen, P. E., Lupwayi, N. Z., Clayton, G. W. 2001. Field comparison of pre-inoculated alfalfa seed and traditional seed inoculation with inoculant prepared in sterile or non-sterile peat. Commun. Soil Sci. Plant Anal. 32(13-14): 2091-2107. Sparrow, S. D. 1988. Inoculation of alfalfa in Alaska. Agroborealis 20(1): 38-40. Sparrow, S. D., Cochran, V. L., Sparrow, E. B. 1995. Dinitrogen fixation by seven legume crops in Alaska. Agron. J. 87: 34-41. Vance, C. P., Heichel, G. H., Phillips, D. A. 1988. Nodulation and symbiotic dinitrogen fixation. Pages 229-258 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. 1st ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA. Vanderkley, J. 2005. Personal communication 2005. Discussion of rhizobia strains. Walley, F. L., Tomm, G. O., Matus, A., Slinkard, A. E., van Kessel, C. 1996. Allocation and cycling of nitrogen in an alfalfa-bromegrass sward. Agron. J. 88: 834- 843.   41 Chapter 3 Subarctic Nitrogen Fixation in a Binary Mix Alfalfa/Smooth Bromegrass and Alfalfa/Timothy Forage Sward 2  3.1 Introduction  Yukon forages are supplied to the local horse and, more recently, the growing cattle industry. The traditional Yukon grass-hay system is one-cut smooth bromegrass (Bromus inermis Leyss.) or, to a lesser extent, timothy (Phleum pratense L.) dominant hay. Harvest occurs in July often with free range post-harvest grazing. Under optimum conditions, with adequate irrigation and N fertilizer rates of 190 kg/ha, the resulting grass hay yields are in excess of 10 t/ha. Background nitrate levels are below 4 ppm in most south central Yukon soils. In order to supplement soil N to achieve optimum plant growth, fertilizers are imported from southern jurisdictions at great cost. In 2008, the cost of 1000 kg of fertilizer landed in Whitehorse, Yukon was approximately $1,100 CAN.  Climatic conditions are conducive to cool season crops. Over 67% of the cropped lands in the Yukon are in forage, with 90% of these lands in monoculture or smooth bromegrass dominant hay.  Planting alfalfa with grasses when a field is first put into production has been common practice in the Yukon. At planting, fields are often fertilized with up to 190 kg/ha N, eliminating any opportunity for rhizobium/alfalfa symbionts to start fixing N. The N ions in the soil solution cause the alfalfa to reduce root exudation of flavonoids, which are critical to the first stage of the infection thread of the bacteria into alfalfa root hairs (Vance et al. 1988). On the other hand, without heavy fertilization the grass stand does not establish full cover and does not aggressively outcompete weeds for nutrients and light. This can lead to a weedy stand and sparse forage establishment.  In this research alfalfa was planted into an existing grass stand with a no-till seeder after the grass was well established and weed populations were less of a  2 A version of this chapter will be submitted for publication. Ball, M.T.A. and Bomke, A.A. Subarctic Nitrogen Fixation in a Binary Mix Alfalfa/Smooth Bromegrass and Alfalfa/Timothy Forage Sward.   42 concern. The approach also included reducing the rate of N fertilizer for mixed grass/alfalfa stands, and assuring proper inoculants and inoculation.  The objective of this study was to determine the effects of seeding alfalfa (Medicago sativa) cv Peace into existing monoculture forage stands of smooth bromegrass and timothy under a combination of N fertilizer rates and different inoculants on forage yield, N content and N fixation. The hypotheses were that: 1. establishing alfalfa inoculated with E. meliloti isolate NRG-34 into existing grass hay stands with no fertilizer application would increase forage yield and N fixation compared to fertilizer applications of 25 kg N/ha, and 2. inoculation of alfalfa seed with a combined TagTeam® treatment (containing Penicillium bilaii PB-50 and Ensifer meliloti NRG-34) would lead to the highest N fixation rates.  3.2 Materials and Methods  This experiment was carried out in the Yukon during the 2005 and 2006 field seasons at the Yukon Government Research Farm, located near Whitehorse, Yukon.  Weather data was monitored over the two years to put the research in perspective to the overall climate of the south central Yukon. Growing season frosts are frequent, with a frost-free period ranging from 25 – 60 days. The growing degree days at this site averaged 722 over the two years of research. The Yukon is classified as a semi- arid environment with annual precipitation averaging 300 mm/yr, approximately 50% of which falls as snow in the winter and is lost through spring runoff and sublimation (Environment Canada 2008). In 2005 the precipitation and mean air temperatures were above normal with reduced GDD because of an early killing frost. In 2006 the precipitation was below normal, but the mean air temperature was near normal and the growing season, as calculated by the GDD, was higher than normal (Table 3.1).       43 Table 3.1: Weather May to September at the Yukon Government Research Farm  Precipitation (mm) Irrigation (mm) ET (mm)1 GDD2 Mean Air Temp (oC) Frost Free (Days) 2005 171 108 405 659 12.2 25 2006 134 80 344 785 10.6 33 Historical Normals 160 * --- --- 743 ** 10.5 * 36 ** 1. Evapotranspiration is the actual ET determined from an evapotranspiration simulator (Etgage Company) 2. The number of growing degree days (GDD) are calculated beginning the fifth consecutive day of the year with daily mean temperatures above 5 C, and terminated the day of the first killing frost (-2.2 C) occurring after July 15. This killing frost temperature does not need to occur as a daily mean temperature, but rather at any moment of a day. * Historical Normals calculated from Environment Canada 30 Year Normal data ** Historical Normals calculated from long term climate monitoring at the Yukon Government Research Farm  The soils at the site are Orthic Brunisols, developed from fine-textured glacio- fluvial deposits underlain by deep glacio-lacustrine deposits, and are classifed as loam, with an average particle size distribution of 42% sand, 47% silt and 11% clay. The average soil pH is 7.0 and has been trending downwards over time with agricultural activity (Hill et al. 2002). Average soil organic matter is around 2% by weight, typical of the Takhini Valley. The soil is well drained and stone free. These soils belong to the Lewes soil association and are described as having 7.5 to 15 cm of brown loam over 7.5 to 10 cm of yellowish brown non-calcareous loam over calcareous bedded silty clay with good moisture holding capacity (Day 1962; Smith 1990).  The site was cleared in 1987 of willow (Salix spp), aspen (Populus tremuloides Michx.), , spruce (Picea spp), lodgepole pine (Pinus contorta Douglas ex Louden), soap berry (Shepherdia spp), and bearberry (Arctostaphylos spp). Since being cleared, the site has been used intensively for a variety of research projects. The experimental plots are located on a level field surrounded by a dominantly lodgepole pine forest, which provides shelter from winds, but consequently creates greater frost occurrences and a slightly shorter season than is experienced in other large cleared fields of the south- central Yukon. Aside from the frost occurrence, the soil, landscape and climatic  44 properties of the site are typical of those encountered at many farms in the south-central region of the Yukon. The elevation of this site is 660 m above mean sea level.  Alfalfa (M. sativa cv Peace) was direct seeded to a depth of 2 cm at a row spacing of 20 cm using a Truax no-till drill model FLX-812 (2.4 m width) into 2.4 x 5m plots in two year-old stands of either smooth bromegrass or timothy. All Truax drills are equipped with depth bands on all disc openers. The depth bands on the discs provided a 3.5 cm penetration of the blade. The double disc openers create a v-groove in the soil surface for the seed to be dropped into. Seeds dropped into the seed slot about half the depth of the disc penetration, which is ideal for alfalfa (1.5-2.0 cm). Press wheels roll behind the discs and press the seed into the sides of the furrow. The machine is equipped with a small seed box capable of handling small 1- 2 mm alfalfa seed. Seed was delivered at 10 kg/ha into a completely randomized split plot design within adjacent 30 m by 40 m fields that had been seeded to either monoculture Carlton smooth bromegrass or Climax timothy in 2003. Each experimental unit was replicated three times in 2005 and twice in 2006.  Main plot treatments were: 1. Uninoculated alfalfa 2. Alfalfa inoculated with Jumpstart® PB-50 Penicillium bilaii 3. Alfalfa inoculated with N-Prove® NRG-34 Ensifer meliloti 4. A dual inoculation with TagTeam® (NRG-34 + PB-50) 5. Monoculture grass with no alfalfa seeded and 0 or 170 kg/ha/yr N fertilizer.  Each of the main plot treatments 1 to 4 was split, the subplot treatments were 0 and 25 kg N/ha as urea (46-0-0). Alfalfa was inoculated onsite with E. meliloti isolate NRG-34 (N-Prove®), with P. Bilaii isolate PB-50 (Jumpstart®) and a combined TagTeam® inoculant, all from Philom Bios. The N-Prove® formulation had a titre of 3.14e9 rhizobia per gram. The commercially available TagTeam® had an internal titre of 6e8 rhizobia per gram and 2.4e7 Penicillium per gram. In order to provide equal amounts of rhizobia per treatment the application of TagTeam® was increased to equal the higher titre of the N- Prove®. Trials were hand fertilized at seeding with the desired N rate (0 or 25 kg N/ha as urea), KCl (0-0-62) at 30 kg/ha and B (Borax) at 1 kg/ha. Phosphorus levels were already high and no additional P was added. These treatments were compared to a  45 “historical” control (treatment 5), in which the monoculture grass plots were not seeded with alfalfa and either unfertilized or fertilized with 170 kg N/ha.  Yukon mean soil temperatures from 0-10 am remain below 18oC throughout the growing season. Peat-based TagTeam® inoculum from Philom Bios contained Ensifer meliloti isolate NRG-34 combined with Penicillium bilaii isolate PB-50 (previously known as P. bilaji) was used for this research. A study was carried out by Rice, Olsen and Collins in 1995 which compared 226 E. meliloti (previously known as Sinorhizobium meliloti) isolates. Of these isolates, NRG-34 was identified as the fastest growing rhizobia at low temperature (10oC) in a greenhouse, as well as having the most effective and fewest ineffective nodules, the highest nodule fresh weight, and the third highest production of plant dry matter at 12oC in the field (Rice et al. 1995).  The cultivar Peace was chosen for these trials because of its Brown Root Rot (Phoma sclerotioides) resistance, northern hardiness, high yield and positive reputation with Yukon producers.  The total plant N difference method (ND) was used for the determination of N fixation. This method determines N fixation by subtracting the total amount of plant N from an appropriate non-nodulating reference crop. A timothy perennial monocot reference crop was used in this research. The quantity of NF was assumed to equal the difference between N accumulated in the fixing crop and the reference crop (Danso 1995; Ledgard and Steele 1992).  Nitrogen use efficiency was calculated to determine the N recovery (RE) of the grasses over the years using the following equation (Greenwood and Draycott 1989): RE = (Uf – Uo)/Na x 100 Uf is the fertilized N uptake Uo is the unfertilized N uptake Na is the nitrogen applied  The crop was irrigated to meet crop water demand based on evapotranspiration measurements. Approximately 110 mm of water was applied in 2005 and 80 mm in 2006. Vegetation sampling occurred in mid-August in both years. In 2005, the smooth  46 bromegrass was sampled at the anthesis stage and alfalfa was at early bloom with a mean height of 30 cm. In 2006, the smooth brome was harvested at the early anthesis stage and alfalfa was harvested at 15% bloom for the second year crop. Timothy was sampled at the anthesis stage and alfalfa mean height ranged from 30 cm for establishment year plants and 80 cm for second year plants. Alfalfa was harvested at early bloom in the establishment year and 40% in the 2006 second year. Above ground biomass was harvested from within two square meter sampling squares in the center of each plot, clipped with shears at 5 cm height. Each vegetation subsample consisted of above ground biomass, plant counts, heights, stages of growth, percent basal cover, and weed incidence. Results were analyzed based on dry matter yield, N content, and N fixation.  In 2005, the timothy site was sprayed with 2-4D Amine at 60 mL/500 L prior to alfalfa emergence, to reduce the competition from shepherd’s purse (Capsella bursapastoris L.) and narrow-leaved hawksbeard (Crepis tectorum L.). Foxtail barley (Hordeum jubatum L.) was hand picked in both years to minimize competition.  All data were analyzed using a split plot analysis of variance model in SAS® 9.1 (SAS Institute). Means were compared using P=0.05 and the conservative Bonferroni multiple comparisons method.  Bonferroni testing is a stepwise, sequentially rejective method for hypothesis testing.  3.3 Results  Effect of inoculum was significant across alfalfa dry matter and N yield data in the analysis of variance (Appendix D) in both the smooth bromegrass and timothy stands in the establishment years. In the smooth bromegrass/alfalfa stand the plots inoculated with NRG-34 (TagTeam® and N-Prove®) produced an average 200 kg/ha greater biomass and at minimum three kilograms per hectare more N in the establishment year than the plots not treated with NRG-34 (Jumpstart® and control) (Tables 3.2 and 3.3). In the second year the controls became nodulated and the effect of inoculation was eliminated. These differences were difficult to determine not only because of infection of the controls, but also because of low alfalfa yields (Table 3.4).   47 In the 2005 establishment year alfalfa within the timothy stand that was inoculated with NRG-34 produced up to 1.3 t/ha greater biomass than the uninoculated plots (Table 3.6). Yields in the NRG-34 inoculated plots were only slightly greater than the uninoculated plots in the 2006 establishment year, averaging 150 kg/ha greater biomass (Table 3.7). For 2006, the second year timothy and alfalfa stand the NRG-34 inoculated plots yielded 4.5 to 5.5 t/ha combined biomass with 2 to 3.9 t/ha from the alfalfa component (Table 3.8).  Fertilization increased grass biomass and N yield of both smooth bromegrass and timothy in both years and reduced alfalfa biomass and N yield.  In the smooth bromegrass stand mean grass dry matter yields were between 0.5 to 2.3 t/ha higher with the addition of 25 kg N/ha (Tables 3.2 to 3.4). Smooth bromegrass total DM was over 10 t/ha in both years for the “historical” yield treatment with 170 kg/ha N, far exceeding the DM yield of any other treatment.  Within the smooth bromegrass, alfalfa DM yields ranged from 0.1 to 0.4 t/ha in the establishment year and 0.4 to 0.9 t/ha in the second year under unfertilized conditions and with the addition of NRG-34 inoculum. The contribution of alfalfa to the DM weight never exceeded 1 t/ha/yr when mixed with smooth bromegrass. Alfalfa DM weights for the control and P. bilaii (Jumpstart®) inoculated treatments were very low in the establishment years ranging from 7 to 28 kg/ha.  The mixed alfalfa and smooth bromegrass stand with either 0 or 25 kg N/ha provided less than 50% of the forage dry matter yield as from the 170 kg/ha N fertilizer treatment.  48  49  Nitrogen content was assessed for each treatment based on the DM weights combined with the N in the above ground biomass. Alfalfa contributed more to the stand’s N content than to its DM yield as the alfalfa provides, per kilogram of vegetation, more N to the stand than a graminoid. Increased alfalfa content in a hay creates a more proteinaceous forage. In the establishment year and second year, alfalfa contributed a maximum of 13.6 kg N/ha and 21.8 kg N/ha, respectively, to the smooth bromegrass mixed forage sward. Mean alfalfa N yield was generally higher with the unfertilized treatments, than with the 25 kg N/ha fertilizer level, the fertilizer treatment often had a significant impact on the dry matter yield of the alfalfa (Table 3.5).  Table 3.5: Analysis of Variance of 2005 Dry Matter Alfalfa Yield in the Seeding Year Source                 DF        F Value       Pr > F  Inoculum               3        21.36     0.0004  R(Inoculum)            8         2.44     0.1147  Fertilizer             1       11.11     0.0103  Inoculum*Fertilizer    3         4.7    0.0355   Alfalfa, in the second year of the mixed smooth bromegrass stand, contributed a greater amount of N to the sward than in the first year, upwards of 20 kg/ha.  The monoculture timothy grass yields were much lower than the smooth bromegrass, with between 6.5 and 7.2 t/ha/yr DM (Tables 3.6 to 3.8). In the 2005 establishment year, the 0 N N-Prove® and 0 N TagTeam® treatments provided 1.4 t/ha and 1 t/ha respectively. In the 2006 establishment year, the alfalfa DM weights were lower, with a maximum of 0.4 t/ha for the 25 N TagTeam® treatment. In the second year the mean DM weight of the 25 N N-Prove® and 0 N Control treatments were only 1.8 t/ha below the heavily fertilized “historical” timothy yield treatment with 170 kg/ha N.  50  51  In the establishment year data, N content in the monoculture timothy stand was greater in all cases, with between 70-107 kg more N/ha. The alfalfa yields of the N- Prove® and TagTeam® treatments with no fertilization contributed substantially to the sward in the establishment year. In year two the contribution of the alfalfa became dominant in some of the experimental units. Three unfertilized samples (0 N TagTeam®, 0 N N-Prove®, and 0 N Control) provided N yields on a per hectare basis that are equal to, or greater than, the timothy stand fertilized at 170 kg N/ha. The N yield of the alfalfa in the 0 N N-Prove® treatment was 102 kg/ha, equivalent to 90% of the total N yield of the mixed sward.  Nitrogen fixation rates were determined using graminoid reference crops.  Table 3.9: Mean Non N-Fixing Reference Value Smooth Bromegrass  2005 2006 Mean (kg/ha) 21.6 26.4 SD (kg/ha) 2.3 1.9 CV (%) 10.8 7.3  Table 3.10: Mean Non N-Fixing Reference Value for Timothy  2005 2006 Mean (kg/ha) 18 23 SD (kg/ha) 4.7 4.6 CV (%) 26.1 19.9  Based on the monoculture timothy reference crop, alfalfa in the mixed timothy/alfalfa stand provided mean N fixation rates of upwards of 35 kg/ha in the establishment year from the N-Prove® treatment with no fertilizer. In the second year mean NF rates are over 90 kg/ha from the same treatment. The most startling trend is the difference in the NF rates in year two based on the addition of only 25 kg/ha N (Table 3.11). Adding this limited amount of N fertilizer to the NRG-34 inoculated treatments (both the TagTeam® and N-Prove® treatments) reduced the NF of the sward by an average of over 20 kg/ha.   52   Table 3.11: Analysis of Variance of Nitrogen Fixation Rates of Alfalfa in Timothy in Year 2 Source                 DF     F Value     Pr > F  Inoculum               3      5.97        0.0194  R(Inoculum)            8      1.43       0.3128  Fertilizer             1     31.35       0.0005  Inoculum*Fertilizer    3      1.12         0.3961   Nitrogen recovery efficiency was calculated for the “historical’ yield at 170 kg/ha N treatments.  Table 3.12: Smooth Bromegrass N Recovery Efficiency 2005 and 2006  2005 2006 Mean  87.2 40.3 SD  17.5 1.6 CV% 14.1 14.1  Table 3.13: Timothy N Recovery Efficiency 2005 and 2006  2005 2006 Mean  62.9 45.5 SD  6.3 6.4 CV% 10 14.1  The N recovery efficiency varied greatly between years, likely due to differences in the growing degree days, temperatures, precipitation and irrigation. Crop growth response to N fertilizer changes year over year because of variations in the soil N supply and climatic conditions (Belanger 1997). The recovery of fertilizer N in the timothy crop ranged from 63 - 45% in 2005 and 2006 respectively.  The N fertilizer recovery for the monoculture smooth bromegrass stand varied greatly over the two years. The smooth bromegrass recovery of applied N in 2005 was 87% but dropped down to 40% in the second year.    53 3.4 Discussion  Inoculation increased DM yield and N content of alfalfa in the establishment year. Based on observations of the roots of the E. meliloti  treated alfalfa (TagTeam® and N- Prove®) were infected early in the establishment year, unfortunately the rhizobia spread and also infected adjacent control and Jumpstart® treatments within the first year. Therefore, comparison between inocula is not possible in the second year as control plots show elevated DM and N contents, yields similar to those found in the NRG-34 inoculated treatments.  Combining the inocula in the TagTeam® treatment (PB-50 and NRG-34) did not yield greater alfalfa DM, N content, or N fixation than NRG-34 alone. High soil test levels of available P at the study site may have reduced the utility of the P. bilaii (Jumpstart®) inoculum. The low rate of N fertilization, 25 kg/ha, had the most pronounced effect on alfalfa and grass variables a somewhat striking result considering the low N supplying power of the soil at the experimental site.  Smooth bromegrass, a competitive, cold tolerant, rhizotomous, allelopathic (Millar 1992) species, is the dominant hay crop in the Yukon. It was optimistic to seed alfalfa into such a crop and expect good alfalfa yields. Planting a mixed alfalfa and grass sward leads to competition between the grass and alfalfa for light, nutrients and water. As has been noted in other Yukon research, alfalfa does not compete well with bromegrass (White 1993), this research arrived at the same conclusion. In previous Yukon alfalfa trials, N has been added at recommended soil test levels (White 1993) when the alfalfa is combined with smooth bromegrass. By applying N to the soil, abundant nitrate and ammonium ions are available for crop absorption. Established smooth bromegrass, with larger, more rhizomatous rooting systems, will absorb the ions more readily, inducing faster growth, and shading the alfalfa throughout the growing season, thereby limiting alfalfa growth. According to Chan and McKenzie (1971), under the shading conditions characteristic of seedling environments in companion crops, alfalfa seedlings accumulated less dry matter than when unshaded. In companion planting in soil with high N rates, legumes will be outcompeted for soil nutrients and light, and will decline over time in the stand. The N ions in solution also cause a direct effect on NF, with increased N ions in soil solution there is a reduction in alfalfa root exudation  54 of flavonoids which are critical to the first stage of the infection thread of the bacteria (Vance et al. 1988), since the alfalfa are able to absorb available N without having to use energy to synthesize the N.  Nitrogen fixation rates were low in the smooth bromegrass stand. In the establishment years (2005 and 2006) the NF rates of less than 14 kg/ha/yr were achieved. In year two, the N fixation rates ranged from 2 to 13 kg/ha/yr. The combination of patchy establishment, allelopathic response, competition and shading from smooth bromegrass resulted in low biomass production from the alfalfa in all treatments applied to the bromegrass stand.  Some negative N fixation values were recorded in the alfalfa establishment years in the mixed smooth bromegrass/alfalfa stand (Table 3.3), indicating that the reference crop had a higher content of N at harvest than the treatments. The unfertilized monoculture grass reference crops each had different characteristics. Timothy, a bunch grass, is not as competitive as smooth bromegrass in terms of fine root development and nutrient acquisition, especially in the alkaline and dry conditions faced at this site. Nonetheless the mean reference values are comparable (Tables 3.9 and 3.10) and are indicative of the low N supply, 18 to 26 kg/ha, of the soil at the study site.  The rate of NF using the ND method is assumed to equal the difference between N accumulated in the fixing crop and the reference crop (Danso 1995; Ledgard and Steele 1992). According to other research, assuming similar N uptake characteristics for the fixing and non-fixing crops, that the two crops absorb soil N with the same efficiency, may sometimes be inaccurate. Because grasses often use soil N at higher rates than legumes, the use of the ND method can lead to underestimated, sometimes negative values, of NF when using a graminoid reference crop (Carlsson and Huss-Danell 2003; Danso 1995; Hardarson and Danso 1993; Ledgard and Steele 1992), as was shown in this research. On the other hand, the alfalfa N fixation rates in the smooth bromegrass stand are small and may be difficult to determine with the ND method.  In the second year of the smooth bromegrass stand we observed a quantifiable positive contribution from alfalfa to stand N levels with mean NF rates of 12 kg/ha. This demonstrates a trend in the expected direction; however, this is a limited data set and  55 without more years of data, it is hard to make inferences about the stand composition and the contribution of alfalfa in subsequent years.  Seeding alfalfa into a timothy stand benefited the resulting mixed sward. Nitrogen fixation rates in the mixed timothy and alfalfa stand reached 35 kg/ha in the establishment year and 102 kg/ha in the second year. In the Delta Junction area of Alaska, which has a similar climate to the Yukon, NF rates in monoculture stands were around 20 kg/ha/yr without irrigation (Panciera and Sparrow 1995).  Work in Saskatchewan yielded NF rates of 62 – 83 kg/ha/yr in a mixed meadow bromegrass/alfalfa stand and in year two NF rates of 132 – 156 kg/ha/yr (Walley et al. 1996).  Recovery efficiency of fertilizer N dropped dramatically in 2006 for both monoculture stands. Even though ET was lower in 2006, the amount of precipitation and irrigation levels were also lower, and likely reduced the N use efficiency. Nitrogen yields of the timothy were quite low. It is hypothesized that this is due to the high pH, low soil N and low soil moisture conditions. Nitrogen deficiency strongly affects timothy shoot growth and leaf area index (Belanger 1997). The ability of the timothy to recover fertilizer N is suspect and likely due to the high pH and low soil water conditions at the site.  There were fluctuations in the colour of the grass in the unfertilized plots over the seasons, at times showing signs of severe N deficiency in the 0 kg/ha N plots, with chlorosis of leaves. Under low N conditions the ability of the timothy to grow was reduced, leaving opportunity for the alfalfa to fill in the canopy. Research from southern Canada has shown significantly reduced timothy regrowth in a second harvest of the second year (Burity et al. 1989). It is possible that, in this study, reduced harvest effects occur in a subsequent year because of the short season. Timothy yields in this research were around 7 t/ha, which is comparable to timothy yields under N fertilizer conditions in Alaska, which averaged 7.3 t/ha near Anchorage (Sparrow and Panciera 2000). These yields are also comparable to yields at Swift Current, Saskatchewan where timothy yields over 7 t/ha (Belanger 2001) have been reported. Timothy is an important forage grass grown in cool season climates of Europe, Russia and North America. It is a preferred grass hay for horse owners in the Yukon, primarily due to the pleasant smell, high palatability, and having higher Ca than other grass forages.  56   Alfalfa DM was correlated against NF for the second year timothy and smooth bromegrass swards. This enabled the determination of a predictive equation for using Figure 3.1: Scatterplot of Dry Matter of Alfalfa vs Smooth Bromegrass  Figure 3.2: Scatterplot of Dry Matter of Alfalfa vs Timothy  57 DM as the reference for total NF to allow for on-the-ground quantification of NF in field. Alfalfa NF in a mixed timothy stand = 0.0229DM + 2.33 (R2=0.96). Alfalfa NF in a mixed smooth bromegrass stand = 0.0192DM + 1.55 (R2=0.74). An equation based on DM to determine NF, developed by Carlsson and Huss-Danell (2003) for northern Sweden for Medicago sativa, was NF = 0.021DM + 17. The equations derived from this research are similar to that from Carlsson and Huss-Danell (2003), but cross the Y-axis at a lower value and have different slopes. Further research is required to test these equations before they can be used as predictive equations for N fixation based solely on DM.  Scatterplots were generated to determine the effect of the grass population on the alfalfa yields, based on DM.  The expectation is that the greater the grass population (based on DM yield) the lower the expected alfalfa DM. The results are unclear for the mixed alfalfa smooth bromegrass stand, which does not show any clear clustering, though the highest alfalfa yields are with low grass weights (Figure 3.1). The results for the mixed alfalfa and timothy stand do point to reduced alfalfa yields in samples with greater grass DM yield. The clustering of the points in the upper left hand area of the scatterplot demonstrate the greater alfalfa growth occurs under low grass yields (Figure 3.2).  Alfalfa growth was the greatest in the plots where no N was added and also where the grass was not as dense. The expectation when incorporating alfalfa into a pre-existing grass stand is that the greater the grass population the lower the expected alfalfa DM. There is evidence of this in the scatterplots as the higher alfalfa yields are dominantly on the left hand side of the figures where there is reduced grass growth.   58          Figure 3.3: Nitrogen Contribution Comparison of Alfalfa and Grass in 2005 and 2006  Throughout the two study years alfalfa growth was poor in the smooth bromegrass stand (the maximum alfalfa DM yield average per plot was 0.9 t/ha). The concern with using an established smooth bromegrass stand was that this competitive, cold tolerant graminoid is able to accumulate greater N reserves over the season and will outcompete the seedling alfalfa for light, water and soil nutrients. Another concern with bromegrass is the allelopathic response shown when root exudates are applied to germinating alfalfa seed resulting in greatly reduced germination (Millar 1992), a response that is not clearly shown in this research. This potential phytotoxicity in the soil could lead to reduced growth rates on a field scale.  The contribution of the alfalfa component to stand forage N yield increased in the second year, especially in the timothy stand (Figure 3.3). In almost all cases, the addition of 25 kg/ha of fertilizer N led to reduced contribution from the alfalfa. In the  59 establishment year the contribution of alfalfa in the smooth bromegrass stand reached 37% of the total stand N yield in the 0 N fertilized TagTeam® inoculated plots. In the second year, smooth bromegrass stand the alfalfa contribution increased to 50% on three 0 N treatments. During the establishment year the contribution of NRG-34 inoculated alfalfa to 0 N fertilized timothy/alfalfa exceeded 50%, and in year two the 0 N treatments reached almost 90% of the stand contribution. In all cases where NRG-34 inoculant had been applied and no N fertilizer added we see greater N contribution from the alfalfa. The recommendation in extension publications is for a small addition of N in mixed legume/grass forage stands. Based on this research it would seem prudent to keep in mind the species composition, the incorporation of alfalfa after stand establishment, and the conditions during stand establishment.  3.5 Conclusion  Ensifer meliloti isolate NRG-34 was effective in nodulating Peace alfalfa and resulted in N fixation rates of over 90 kg/ha in a first year combined alfalfa and timothy stand. The addition of Penicillium bilaii to the NRG-34 inoculant had no discernable effect on the nitrogen fixation or yield of the alfalfa.  The addition of a limited amount of N fertilizer at 25 kg/ha simulated grass growth depressing yields of the alfalfa. As has been shown in other research (Vance et al. 1988), adding N fertilizer to the stand decreased the fraction of stand N derived from NF.  The bromegrass was not a good companion crop for the interseeding of alfalfa as bromegrass maintains a high foliar density providing a challenge for alfalfa growth. Throughout the two study years alfalfa growth was low in the smooth bromegrass stand with the maximum alfalfa DM yield averaged per plot of 0.9 t/ha. Alfalfa in the timothy mix had much more vigorous growth. Within the timothy stand the alfalfa contributed 50% of the stand N within the first year and over 90% by year 2 providing 3.9 t/ha of DM.  This research highlights the importance of careful N management and grass species selection during alfalfa forage establishment.  60 3.6 References  Belanger, G. 1997. Growth analysis of timothy grown with varying N nutrition. Can. J. Plant Sci. 77(3): 373-380. Belanger, G. 2001. Improving the nutritive value of timothy through management and breeding. Can. J. Plant Sci. 81(4): 577-585. Burity, H. A., Ta, T. C., Faris, M. A., Coulman, B. E. 1989. Estimation of nitrogen fixation and transfer from alfalfa to associated grasses in mixed swards under field conditions. Plant and Soil 114(2): 249-255. Carlsson, G. and Huss-Danell, K. 2003. Nitrogen fixation in perennial forage legumes in the field. Plant and Soil 253(2): 353-372. Chan, W. and MacKenzie, A. F. 1971. Effects of shading and nitrogen on growth of grass-alfalfa pastures. Agronomy Journal 63(5): 667-669. Danso, S. K. A. 1995. Assessment of biological nitrogen fixation. Fertilizer Research 42: 33-41. Day, J. H. 1962. Reconnaissance soil survey of the Takhini and Dezadeash Valleys in the Yukon Territory. Research Branch Canada Department of Agriculture, Ottawa, Ontario, Canada. 53714-2: 1-78. Environment Canada. 2008. Canadian climate normals 1971-2000. Government of Canada, Ottawa, Ontario, Canada. www.weatheroffice.gc.ca. Greenwood and Draycott 1989. Experimental validation of an N-response model for widely different crops. Fert. Res. 18:153-174. Hardarson, G. and Danso, S. K. A. 1993. Methods for measuring biological nitrogen fixation in grain legumes. Plant and Soil 152(1): 19-23. Ledgard, S. F. and Steele, K. W. 1992. Biological nitrogen-fixation in mixed legume grass pastures. Plant Soil 141(1-2): 137-153. Millar, D. A. 1992. Allelopathy in alfalfa and other forage crops in the united states. Pages 169-178 in: Rizvi, S. J. H. and Rizvi, V., eds. Allelopathy: Basic and applied aspects. First ed. Chapman & Hall, London, England. Panciera, M. T. and Sparrow, S. D. 1995. Effects of nitrogen fertilizer on dry matter and nitrogen yields of herbaceous legumes in interior Alaska. Can. J. Plant Sci. 75: 129- 134. Rice, W. A., Olsen, P. E., Collins, M. M. 1995. Symbiotic effectiveness of Rhizobium meliloti at low root temperatures. Plant and Soil 170: 351-358.  61 SAS Institute. 2003. The SAS system for Windows. Release version 9.1. SAS Institute, Cary, NC. Smith, C. A. S. 1990. Nature of the cryic thermal regime of agricultural soils in the Yukon Territory, Canada. In Proceedings International Symposium on Frozen Soil Impacts on Agricultural, Range and Forest Lands 90-1: 11-20. Sparrow, S. D. and Panciera, M. T. 2000. Forage yield and soil characteristics under various crops in Alaska. Acta Agric. Scand. 50(2): 75-81. Vance, C. P., Heichel, G. H., Phillips, D. A. 1988. Nodulation and symbiotic dinitrogen fixation. Pages 229-258 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. First ed. ASA-CSSA-SSSA, Madison, Wisconsin. Walley, F. L., Tomm, G. O., Matus, A., Slinkard, A. E., van Kessel, C. 1996. Allocation and cycling of nitrogen in an alfalfa-bromegrass sward. Agron. J. 88: 834- 843. White, M. 1993. Yukon agricultural research and demonstration report 1992/93. PR 94- 3: 1-10.  62 Chapter 4  Conclusion  By understanding how to establish alfalfa in the Yukon, we can realize the benefit of N fixation, using the inert N in the atmosphere to increase plant available N in Yukon farm systems. This helps reduce the need for imported N based fertilizers to supplement soil N, leading to a reduction in the expense of buying and transporting N fertilizer, thereby translating into a lower cost of production for farmers. This also reduces the impact on the environment, and ultimately creates a more sustainable farm in the subarctic North.  This study focused on alfalfa as one of the most winterhardy and cold tolerant of the agronomic legumes (McKenzie et al. 1981). Medicago sativa cv Peace, which had been grown with limited success in previous Yukon research (Bisset 1989; White 1993), was seeded in monoculture and mixed smooth bromegrass and timothy stands. It was seeded in combination with bacterial and fungal inoculants which were previously untested in the Yukon. Irrigation and fertilizers were added to provide a healthy soil environment for the establishment of rhizobia/alfalfa symbiosis.  4.1 Comparison of Research Results  Some of the results of this research met the expectations of the hypotheses. In all cases the inoculation with NRG-34 (both in the TagTeam® and N-Prove® formulations) provided high nodule numbers and N fixation rates. This inoculum, combined with the unfertilized N treatments, provided the highest dry matter yield, N content and N fixation results for alfalfa in all trials. Nitrogen fixation rates in year two ranged from 57 to 90 kg/ha/yr for the NRG-34 inoculated treatments in a timothy-alfalfa sward.  The unfertilized treatments sometimes achieved higher N fixation rates in year one and all except the smooth bromegrass TagTeam® treatment achieved higher rates in year two. In year two, the addition of only 25 kg N/ha in the timothy and alfalfa stand led to differences of between 20 to 50 kg/ha/yr N fixed. Under the N fertilized conditions the grass growth was greater, yielding between 1 and 2 t/ha more biomass, competing with the alfalfa for light, moisture, and soil nutrients.   63 In the monoculture alfalfa trial, the yields per year were in the range of 3 t/ha. This yield is less than the alfalfa in the mixed timothy stand, which yielded upwards of 3.9 t/ha. Within the mixed alfalfa and timothy stand the unfertilized, N-Prove® treatment yielded the highest N fixation at 90 kg/ha/yr, also higher than in the monoculture stand. This is surprising considering the alfalfa in the mixed stand had to compete for nutrients and water against the timothy. Differences between the two research plots may have contributed to this result. The amount of irrigation in the monoculture stand was slightly lower than in the mixed stands. The vegetation cover of the timothy may have maintained better moisture conditions, slowing evaporation from the soil surface. As well, transpiration may have been limited by the poor timothy growth. There was also a difference in the seeding rate between the stands, that of the monoculture being slightly higher than in the mixed stands, which may have led to competition among the seedlings, resulting in lower yields. Research by Ta and Faris (1988) in growth chambers and Burity et al. (1989) in the field showed increased alfalfa NF in mixed stands. Ta and Faris (1988) hypothesized that the release of root exudates by timothy might result in this enhanced NF.  Poorest alfalfa yields were observed when alfalfa was seeded into smooth bromegrass. Throughout the two years there was no substantial alfalfa growth in the smooth bromegrass stand. Seeding alfalfa into smooth bromegrass was always considered a risk as it is an allelopathic (Millar 1992), cold tolerant graminoid which has been known to out-compete seedling alfalfa for both light and soil nutrients (Chan and MacKenzie 1971) and is competitive in other northern environments (Sparrow and Masiak 2004).  In the monoculture alfalfa trial late season harvesting reduced the alfalfa yield by half in the second year. In the establishment year the monoculture TagTeam® inoculated alfalfa yielded 3.1 DM t/ha containing 77 kg/ha of NF.  In the second year the unharvested inoculated alfalfa yields were 3.4 t/ha with NF of 66 kg/ha compared to the late harvested stands, which yielded 1.5 t/ha and an NF rate of only 20 kg/ha. The effects of the late season harvest are startling and reflect the importance of removing grazing animals during the fall to allow plant energy reserves to accumulate in the roots.  All of the control plots, which were seeded without inoculum, had nodulation by the end of the first growing season. Movement of the bacteria through the action of soil water and wind likely infected the control plots; an alternate, but unlikely, hypothesis is the infection of these plots by native rhizobia. One of the concerns was the inability to test the DNA of the rhizobia to determine that the root infections were NRG-34 rhizobia. No PCR test was available throughout the time of this research.  64  With reference to the third hypothesis, it was very difficult to discern any differences between the TagTeam® and N-Prove® treatments. The mixed stands at the Yukon Government Research Farm had high soil P levels due to fertilization the year before the trial was initiated (2004) so the effect of the P. bilaii in the soil increasing P availability through the release of P solubilizing acids was not noticeable.  The fourth hypothesis, that the N recovery in the biomass will be higher in a mixed stand with low or no fertilizer N, was optimistic. The grass stands, fertilized with 170 kg N/ha, were provided with abundant N ions in solution resulting in high N uptake and yields. Conventionally fertilized smooth bromegrass contained 169 kg/ha N in 2005 and 95 kg/ha in 2006. Smooth bromegrass stands provided higher total biomass and N contents than the alfalfa or combined alfalfa and smooth bromegrass mixed stands in all cases. In the timothy stand the results are encouraging. The seeding of alfalfa into an existing timothy stand and reducing fertilizer levels to zero resulted in N contents that are slightly higher than the conventionally fertilized timothy stand. The 0 N Control and 0 N N-Prove® treatments provided between 2 and 14 kg/ha more N in the stand than the heavily fertilized monoculture stand.  4.2 Strengths and Shortcomings of Research  It was both a strength and a weakness to work in the field and strive for results that are applicable on the field scale. The variability shown in the data is indicative of what would likely occur in a large field, where it is difficult to control variables. Shortcomings of the research can be divided into 2 categories: design compromises and in-field problems.  Design compromises include few replications and lack of field space. More replications of experimental units would have been desirable at all sites. Unfortunately there was not sufficient space at the Yukon Government Research Farm to replicate the trials further. The size of the plots had to be large enough to allow the no-till drill and tractor to function. In year one, some harvesting occurred near the edges of the plots. It became apparent that this was not a good strategy, especially along the edge where the seeder was starting as the seed flow would not be consistent for at least the first metre.  There were a number of problems throughout the growing seasons. Mule deer developed a taste for the alfalfa and grazed some of the edges of the treatments. Working in this wildlife- rich environment, there are always likely to be problem animals. Exclusion fencing would have  65 been a good option, but it would have required many fences and was not thought of until after the majority of grazing had occurred. Although the deer did eat the treatments, they usually stayed to the outside, eating the buffer area that was not sampled for analysis.  Other field concerns included patchy establishment of the alfalfa, due to a combination of factors including inconsistent seeding with the no-till seeder, varied density of the grass stand in which the seed was being planted, and low soil moisture. When the irrigation data was compared to evapotranspiration measurements at the end of the seasons, it was noted that in both years the irrigation levels were not adequate to replace the ET. In 2005 irrigation and precipitation provided 70% of ET and in 2006 down to 63%.  There were a number of strengths in the research. Although the cooperator trials were considered only moderately successful (Appendix G), research results were shared with local farmers at these sites. Information was also exchanged during field days and in written correspondence, including newsletters and emails to the local farmers. Working cooperatively on farm also provided one of the best methods to disseminate information to the farm community. Having three different experiments at the Research Farm provided useful comparisons. The experiments followed closely on the hypotheses that were initially set out. Determining ranges for N fixation in south central in the other chapters, you wrote this as one word. This version looks correct to me, but be consistent Yukon is a useful result that can be applied to future research and extension work.  A sensible approach to sustainable agriculture is to provide a closed system as much as possible. This study determined that up to 90 kg/ha/yr can be provided on farm by planting alfalfa. With these results it would appear that the combination of animal manure, compost, and legumes can provide all of the N needed for a Yukon farm enterprise, whether it be for livestock production or used in crop rotations. It is encouraging that the N fixation results from this research compare with those provided by Klebesadel (1978), and Sparrow and Panciera (2001) from Alaska.  This research plays a small role in examining production systems that rely on what the natural system has to offer, not on what can be manufactured and applied. Although it may be common sense to use a legume crop as part of the nutrient source to build a sustainable farming system, it does not necessarily create an economically feasible business. The economics must be proven and this research only goes part of the way to show that unfertilized mixed timothy  66 and alfalfa swards are potential cropping options in the North. Further research needs to examine harvesting methods, drying time for alfalfa hay, and palatability and overall hay quality.  Another similar study has already been initiated focusing on the use of legumes and organic fertilizers to improve soil nutrient levels, microflora growth and plant yields. Six different legumes were seeded in different locations around the Yukon. To date, the yields of alfalfa and sweet clover, are the most encouraging.  A portion of the research was not reported on in the main body of the thesis. Soil temperatures and moisture contents were measured throughout the two growing seasons (Appendix A) to understand some of the environmental conditions in the area. Work was also conducted investigating N transfer in the mixed smooth bromegrass stand using PRS™ probes was interesting, but contributed little to the our understanding of NT from legumes to grass in the mixed stands (see Appendix F).  4.3 Future Research  The future holds many opportunities for growing crops in the subarctic North, with the rise in the costs of both chemical fertilizers and the transportation of goods, it will become even more critical to examine solutions for reducing chemical and imported fertilizer applications. There are three options available for reducing imported fertilizers application: use locally available alternatives; use less fertilizer by better understanding nutrient mobility and change; and develop symbiotic plant-microbe relationships.  In the development of the research forming this thesis, some time was spent examining locally available sources of nutrients that could be used as fertilizer: city compost, fish wastes and human wastes. However, there is very limited supply of each of these nutrient sources and transportation costs could limit their geographic distribution as compared to incorporation of legumes into farming systems. Nonetheless, more work is needed in these areas. Using less chemical fertilizer by researching precision fertilizing practices and different fertilizer blends is an important project that should be undertaken at some point in the Yukon.  Continued research into agronomic legume crops and accompanying N fixation rates in subarctic conditions is needed, including long term studies of alfalfa growth and management, and better research on N transfer in subarctic mixed stands.   67 There are a number of native N fixers, both leguminous and non-leguminous that may be used in agronomic systems. The N fixation rates and management of soapberry (Shepherdia canadensis L.), alder (Alnus sp.) and lupines (Lupinus arcticus S. Watson) need further research and study. Lupines and other native leguminoserums of the Oxytropis and Astralagus genus are possibilities as soil builders in an agricultural system. The other species, with woody bark, are not compatible with crop rotation or foraging, but can be used as indicators of where to develop new farm lands (Klebesadel 1978). In alder groves or where there are concentrations of soapberry bushes, the soil N levels should be greater than the surrounding soils. Native legumes provide a small amount of N. Under inoculation and management, these rates may improve substantially.  A number of agronomic legumes offer some promise, including sweetclover (Melilotus alba and M. officinalis), which offers great potential as a biennial that will grow in very low nutrient soils and produce abundant shoot biomass. Sweetclover also has the ability to colonize disturbed terrains as a primary succession plant and promote microbial activity in these conditions. White clover (Trifolium repens), red clover (Trifolium pratense), alsike clover (Trifolium hybridum), and hairy vetch (Vicia villosa) need further study, as well as straight or intercropping annual legumes such as pea (Pisum sativum L.). Further alfalfa varietal assessments also need to be conducted.  It is important to keep in mind that the limiting nutrient in northern systems is almost exclusively N; therefore, to produce crops, a small amount of chemical fertilizer N is an option to balance out soil nutrition. Using available nutrient sources in appropriate amounts is a good tool for farmers to use in order to supply the needed nutrients to the soil.  Future research should examine plant available N in subsequent years when alfalfa is used as a plowdown. One of the concerns with growing N-fixing crops in rotation in dry, cold, northern environments, is the length of time for the plant materials to decompose and mineralize N for subsequent crops. In rotation, the N is available from the death of and plowing in of, shoot and root biomass, nodules, and a small portion from exudation (Burity et al. 1989; Heichel and Henjum 1991). In work by Sparrow and Panciera (2001), the amount of N available from an alfalfa rotation for a subsequent dryland barley crop was limited. For the unfertilized treatments, mineralized N available for the next crop was highest for the alfalfa treatment compared to a grass, grain and fallow treatment; however, this did not translate into the greatest biomass accumulation in the barley (Sparrow and Panciera 2001). It would be worthwhile to determine pathways for N availability in crop rotations in the south central Yukon.  68  More research on the process of interseeding alfalfa into dry rangelands to improve forage production and nutritive value is needed. Throughout Canada, this research has generally been met with limited success due to lack of competitiveness and persistence of the common varieties of alfalfa under free range grazing. Often these studies use some limited amounts of N fertilizer. In grass legume pastures, N fertilization usually produces greater grass growth outcompeting legumes, which results in reduced legume yield (Jacobsen and Surber 1995). And as discussed previously, free range on grazing lands is detrimental to legume growth in northern environments.  Continued work on breeding and rhizobia selection is needed. The work of Danielle Prevost and others at the Agriculture and Agri-Food Canada Soils and Crop Research and Development Centre in Sainte-Foy, Quebec provides some promise of psychotrophic rhizobia that will nodulate agronomic species in cold regions. So far Prevost’s research is limited to sainfoin (Prevost et al. 2003). Field research accumulating native rhizobia in northern soils combined with laboratory selection of rhizobia that are both competitive and able to nodulate at cold temperatures should be followed up in the Yukon.  A considerable amount of the recent work being conducted on alfalfa is focused on genetic modification to create glyphosate resistance for weed control. Weeds were controlled in this research with existing pesticides and manual labour, which resulted in low weed populations in most cases. Weed control was also assisted by the alfalfa itself, which in the monoculture and timothy stands, had expansive shoot growth that competed with other species, therefore reducing their development. There is research showing that glyphosate application reduces the rate of NF (Motavalli et al. 2004; Reddy and Zablotowicz 2003). Therefore, although glyphosate resistance might be a useful weed management technology in the North, it reduces the overall benefit of seeding alfalfa. If it can be shown that weed management is aided and N fixation rates maintained, then a last requirement for seeding this GM seed is that it be absolutely imperative there be zero pollen transfer. It is not ethical to genetically contaminate a neighbouring field or the natural environment.  We must also be aware of the concern of planting cold hardy, N fixing crops in the pristine subarctic environment. These plants are able to survive in the North and can produce their own N. They can therefore survive in low nutrient environments, and can become invasive species. Sweetclover is one such plant that has become just such a concern along disturbed linear corridors (highways and waterways) in the Yukon.  69  4.4 Concluding Statement  Since the Second World War, agri-business has taken farmers in the direction of NPK chemical fertilizer management as the dominant method to help treat soil nutrient deficiencies. This technological advancement enabled farmers to extract from the soil as much productivity per hectare as possible over a short period of time. This is not truly a sustainable system, as the inputs are expensive in terms of energy use, and in the case of the Yukon, are imported from a long distance. This research demonstrated that the most competitive grass, smooth bromegrass, under optimum chemical fertilizer and irrigation conditions, provides a yield greater than any achievable for a legume/grass mixed sward. From an economic perspective the value per acre is greater using a smooth bromegrass hay and fertilizer, yet from a sustainablility perspective, the bromegrass system requires a great amount of energy input, energy that is not required to the same level in mixed legume/grass stands (Pimentel and Pimentel 2007). Smooth bromegrass and fertilizer combinations rely on transportation links and fertilizer production, whereas a mixed legume/grass sward provides an alternative option that, although requiring some fertilizers, does not have the chemical N demand inherent in graminoid fields.  Forty years ago J.Y. Tsukamoto, Superintendent at the Haines Junction 1019 Experimental Farm, questioned the commercial status and viability of farming in the Yukon (Abbott et al. 1960). It is true that the Yukon has a small industry, hampered predominantly by climate. As global warming progresses in the North with increased warming trends forecast for almost the entire Yukon (Bernstein et al. 2007), more agronomic crops will be grown. It is prudent to understand the limitations of the production system we work within and to keep pushing the boundaries of crop feasibility. We ought not close our eyes to the potential that is around the corner. Continued research demonstrating agronomically attractive, economically viable cropping systems that operate in a sustainable farm system is necessary. Furthering our ability to provide as much as possible the foods from our local farms to feed Yukon’s growing population will contribute to a sustainable future for the Yukon.  70 4.8 References  Abbott, J. W., Hough, W. H., Tsukamoto, J. Y., Morrison, J. W. 1960. Experimental Farm Mile 1019, Alaska highway Yukon Territory progress report 1953-1959. 64254-6: 1-20.  Bernstein, L., Bosch, P., Canziani, O., Chen, Z., Christ, R., Davidson, O., Hare, W., Huq, S., Karoly, D., Kattsov, V., Kundzewicz, Z., Liu, J., Lohmann, U., Manning, M., Matsuno, T., Menne, B., Metz, B., Mirza, M., Nicholls, N., Nurse, L., Pachauri, R., Palutikof, J., Parry, M., Qin, D., Ravindranath, N., Reisinger, A., Ren, J., Riahi, K., Rosenzweig, C., Rusticucci, M., Schneider, S., Sokona, Y., Solomon, S., Stott, P., Stouffer, R., Sugiyama, T., Swart, R., Tirpak, D., Vogel, C., Yohe, G. 2007. Climate change 2007: Synthesis report. An assessment of the intergovernmental panel on climate change. 52 pp. Bisset, K. 1989. Yukon crop development program 1989 field season report. Yukon Renewable Resources. 150 pp. Burity, H. A., Ta, T. C., Faris, M. A., Coulman, B. E. 1989. Estimation of nitrogen fixation and transfer from alfalfa to associated grasses in mixed swards under field conditions. Plant Soil 114(2): 249-255. Chan, W. and MacKenzie, A. F. 1971. Effects of shading and nitrogen on growth of grass- alfalfa pastures. Agron. J. 63(5): 667-669. Heichel, G. H. and Henjum, K. I. 1991. Dinitrogen fixation, nitrogen transfer, and productivity of forage legume-grass communities. Crop Sci. 31: 202-208. Jacobsen, J. S. and Surber, G. W. 1995. Alfalfa/grass response to nitrogen and phosphorus applications. Commun. Soil Sci. Plant Anal. 26(7&8): 1273-1282. Klebesadel, L. J. 1978. Biological nitrogen fixation in natural and agricultural situations in Alaska. Agroborealis January: 9-12. McKenzie, J. S., Pankiw, P., Siemens, B. 1981. Peace alfalfa. Can. J. Plant Sci. 61: 473-474. Millar, D. A. 1992. Allelopathy in alfalfa and other forage crops in the United States. Pages 169- 178 in: Rizvi, S. J. H. and Rizvi, V., eds. Allelopathy: Basic and applied aspects. 1st ed. Chapman & Hall, London, England. Motavalli, P. P., Kremer, R. J., Fang, M., Means, N. E. 2004. Impact of genetically modified crops and their management on soil microbially mediated plant nutrient transformations. J. Environ. Qual. 33(3): 816-824. Pimentel, D. and Pimentel, M. 2007. Food, energy, and society. 3rd ed. CRC Press, New York. 380 pp. Prevost, D., Drouin, P., Laberge, S., Bertrand, A., Cloutier, J., Levesque, G. 2003. Cold- adapted rhizobia for nitrogen fixation in temperate regions. Can. J. Botany 81(12): 1153- 1161. Reddy, K. and Zablotowicz, R. 2003. Glyphosate-resistant soybean response to various salts of glyphosate and glyphosate accumulation in soybean nodules. Weed Sci. 51(4): 496-502.  71 Sparrow, S. D. and Panciera, M. T. 2001. Crop yield and soil characteristics following various cropping regimes in Alaska. Acta Agric. Scand. 51(3): 143-150. Sparrow, S. D. and Masiak, D. T. 2004. Forage crop variety trials in the Tanana Valley of interior Alaska. AFES Circular 125: 1-32. Ta, T.C. and Faris, M.A. 1988. Effects of environmental conditions on the fixation and transfer of nitrogen from alfalfa to associated timothy. Plant Soil 107: 25-30. White, M. 1993. Yukon agricultural research and demonstration report 1992/93. Yukon Renewable Resources. PR 94-3: 1-10.   72 Appendix A Experimental Conditions  The Yukon Territory is located in northwestern Canada between 60o and 69o North latitude. Alaska sits to the west, the Northwest Territories to the east, the province of British Columbia to the south and the Beaufort Sea to the north.  Conditions north of 60o are more difficult for farming than elsewhere in Canada. The short, dry growing seasons, mid season frosts, cold soil temperatures, high costs of labour and transportation, and the added challenge caused by undulating permafrost- affected thermokarst topography, combine to create an unforgiving environment. It is a display of brave optimism to observe Yukon farmers clearing aspen (Populus tremuloides Michx.), willow (Salix spp) , and spindly spruce trees (Picea spp) in order to plant crops and build the locally available food supply.  The Yukon is part of the northern extension of the Canadian cordillera characterized by mountainous terrain with broad upland plateaus. The elevated mountainous terrain is dissected by steep to rolling foothills cut by broad, glacially carved valleys. These narrow, interconnected valleys are occupied by large rivers and lakes, and it is only on the benches and uplands near these waterbodies that lands are farmed. Agriculture is limited to below 64o5’ latitude and to the major river valleys including those of the Yukon, Takhini, Pelly, Stewart, and Liard (Hill et al. 2002). The Takhini valley in the south central Yukon, is within the main agricultural region in the Yukon. This area is the most intensively farmed within the Yukon, not necessarily due to a better climate or superior soils, but rather due to the proximity to Whitehorse, the main market for produce sales and for off farm income.  Unforgiving Climate Climate is the major limiting factor for agriculture in the Yukon; the short frost-free period and lack of heat units during the growing season greatly reduce the time for production. The climate of the Yukon is characterized by long, cold winters and short, warm summers. This subarctic continental climate has temperatures reaching as high as 36oC in the summer and as low as -60oC in the winter. The average frost-free period ranges from 93 days in the Watson Lake area (southeast) to 21 days in the Haines  73 Junction area (southwest) (Environment Canada 2008). In addition to the frost-free period having a great geographical variation, there is also substantial temporal variation from year to year at any location.  Limited precipitation is another barrier to crop production in this region. Average annual precipitation ranges from less than 200 mm west of Whitehorse to more than 400 mm in Watson Lake (Environment Canada 2008). The south central Yukon, where this research was undertaken, lies within the rainshadow created by the St. Elias and Coastal mountains. South central Yukon is subject to droughts between April and June (Environment Canada 2008), a particular problem for crop germination under dryland conditions. The bulk of the rainfall occurs in July and August, which can create problems for drying forage crops (Table A.1).  Agroclimatic conditions around Whitehorse are characterized by long day lengths, short summers, and with late July and August precipitation. Long hours of daylight during the summer promote rapid growth that compensates, to some extent, for the cooler summer temperatures experienced north of 60o latitude (Hill et al. 2002). Day length at the Whitehorse airport (60o42’) is 19 hrs 8 min on the longest day in June, with a 30-year average of 269.4 hours of sunshine during June (Environment Canada 2008).  Table A.1: Long Term Climate Averages from May to September in the south central Yukon  May June July Aug Sept Avg day length (hrs) 17.3 19 18.2 15.8 13 Avg sunlight (hrs) 261.5 269.4 249.9 232 135.7 Max available sunlight (hrs) 536.9 569.6 564.1 488.4 388.8 1971-2000 Avg rainfall (mm) 13 29.7 41.4 38.5 29.3 1971-2000 Mean temperature (oC) 6.9 11.8 14.1 12.5 7.1 (Environment Canada 2008)       74 Table A.2: Weather Data from May to September at the Yukon Government Research Farm Year Precip (mm) Irrigation (mm) ET (mm) 1 GDD 2 EGDD 3 Mean Air Temp (oC) Frost Free (Days) 2005 171 108 405 659 778 12.2 25 2006 134 80 344 785 925 10.6 33 1. Evapotranspiration is the actual ET determined from an ET gauge (Etgage company) 2. The number of growing degree days (GDD) are calculated beginning the fifth consecutive day of the year with daily mean temperatures above 5oC, and terminated the day of the first killing frost (-2.2oC) occurring after July 15. This killing frost temperature does not need to occur as a daily mean temperature, but rather at any moment of a day. 3. The GDD calculation is adjusted upward by 18% to account for day length, the boost plants receive from the long hours of daylight north of 60° latitude (Tarnocai et al. 1988).  The growing season, as calculated by the growing degree days, is quite short, usually due to the short time between killing frosts (Table A.2). Based on research from Beaverlodge, Alberta, alfalfa cannot survive freezing temperatures below –5oC in mid summer (McKenzie et al. 1988). The Beaverlodge research was carried out in summer temperatures that were on average higher than those in Yukon. I believe that under Yukon conditions plants are somewhat hardened throughout the growing season, as the mean monthly air temperatures in June and August are barely above 10oC. Although mid summer killing frosts would be assumed to have an effect on plant growth during peak summer temperatures, based on observation these frosts do not stop alfalfa production in the Yukon. Therefore the limit of the season as measured by killing frosts may not be indicative of the full potential growing season for this frost tolerant legume.  The Yukon Government Research Farm is in a small cleared area that causes increased frost occurrence as shown by the short frost free period in table A.2. It is important to locate trials in areas with good air drainage to reduce the effects of frost on plant growth through the growing season.The 2005 growing season began with a warmer than normal spring (April and May) leading to an early start to the growing season. Very little precipitation fell in May and the daily temperatures were above normal. The temperatures remained above average until mid June at which time temperatures dropped to seasonal averages. Precipitation was below the seasonal normal at the Research Farm until the end of June when a record breaking one day  75 rainfall of 27 mm brought the precipitation total for the month above normal at 62 mm. July and August temperatures and rainfall returned to above normal values. The following year, 2006, was much cooler at the beginning of the season, resulting in slower spring growth. Although the monthly temperature in May was still above normal, it was cooler than the previous spring. In 2006, mean air temperatures were below normal for the overall season due to well below normal temperatures in August.  Soil Conditions In the Takhini Valley agricultural soils are formed on glaciolacustrine silt and clay deposits of ancient Lake Champagne (Hill et al. 2002). Soils throughout the south central Yukon are low in organic matter, and salinity has been identified as a problem in some localized areas (Rostad et al. 1977).  Permafrost is found throughout the Yukon varying from sporadic discontinuous in southern agriculture areas and increasing to extensive discontinuous at the northern extreme of agriculture activity in the Yukon (around Dawson City 64oN). This thermokarst topography leads to an additional challenge. It has been noted by some producers that after land clearing, the level of the land subsides in some areas. The loss of the insulation provided by the forest cover, and the subsequent change in the thermal regime causes the ice lenses to melt, leading to undulating terrain, which reduces the accuracy of seeding, fertilizing and harvesting equipment, making crop management more difficult.  The soils around the Whitehorse area are heterogeneous. Within the Whitehorse area there are 19 soil associations (Rostad et al. 1977). In the locations where the trials were located the Brunisols are part of the Lewes or Champagne soil association, with a sandy loam to silt loam texture. These are mineral soils with very little organic matter, and therefore absorb and transmit heat readily. The Lewes soils remain cold in the spring, but by early June the soil temperatures average 10oC based on data collected at the Yukon Government Research Farm (research from this thesis). Soil pH tends to be near neutral to slightly alkaline, with only localized salinity concerns in some depressional sites.   76 Fertile soils help in alfalfa stand establishment, overwintering, persistence, promotion of early growth, increasing yield and quality, and improving disease resistance. Takhini Valley soils are only weakly weathered (Smith 1990) and therefore contain most required micro nutrients in sufficient quantities, but some key macro nutrients are in low supply, particularly N and P. Potassium and S abundance is often dependent on local geology. The most common micronutrient deficiency is B (Hill et al. 2002), a concern for N fixation. Alfalfa production removes a range of nutrients from the root zone which must be replenished (see Table A.3). Nutrient levels are easily monitored with conventional soil tests and amendments can be applied where needed.  Table A.3: Estimated Nutrient Removal in kg/ha per 1 tonne of Dry Matter Alfalfa for Various Regions of North America P 3.6 – 7.4 Ca 13.3 – 37.1 Zn 0.03 – 0.06 P2O5 12.4 – 18.5 Mg 2.3 – 8.7 Cu 0.008 – 0.01 K 30 – 68 Fe 0.13 – 0.37 Mn 0.05 – 0.12 K2O 61.8 – 80.3 B 0.03 – 0.1 Mo 0.0025 S 2.4 – 8.7 (Lanyon and Griffith 1988; Undersander et al. 2004)  The ranges provided in Table A.3 are for varying yields throughout North America. Nutrients are being replenished in the soil system through chemical and microbial activity, but the cold, dry soils in the North slow both the conversion of nutrients and organic matter decomposition, limiting the mineralization of nutrients and the amount of nutrient available for plant uptake.  Nitrogen and P increases in the soil system are expected with the coinoculation of alfalfa seed with the Ensifer meliloti and Penicillium bilaii. With inoculation, N fertilizers are often unnecessary in a mixed legume/grass stand. Research undertaken in Minnesota confirms that NT from legume to grass does occur and that forage legumes in binary mixtures with grass can be self-sufficient for N (Heichel and Henjum 1991). In subarctic climates, with slower decomposition, the rate of NT will be lower, therefore a small amount of N may be necessary for grass health.    77 Soil Temperature Mean, minimum and maximum soil temperatures were collected for the 0-10 cm depth in the soil plow layer at a location near the Rafter ‘A’ Ranch cooperator field trial (Figures A.1 and A.2). Diurnal soil temperatures in the upper soil profile varied substantially through the two year study.  The daily minimum and maximum temperatures ranged up to 10oC within a 24 hour period. The data was recorded using a 10 cm stainless steel temperature probe and HOBO dataloggers (Onset® Corporation).  Minimum soil temperature for NF is 10oC (Rice et al. 1995). Mean daily soil temperatures at the beginning of May in 2005 were already over 10oC (Figure A.1), as 2005 had an abnormally warm spring. In 2006, the soil warmed up much more slowly and only reached a mean of 10oC in late May (Figure A.2), approximately one month later than in 2005. Variability between the minimum and maximum temperatures at the shallow depth shows the diurnal cycle of heating from solar radiation. The soil temperature maximums reached 20oC in early June 2005, spiked upwards through to early July and then recorded another spike in mid-August 2005. In 2006, the soil temperature only reached 20oC once, but exhibited similar diurnal fluctuations with spikes in mid-June through July. The length of time during the season at which the mean soil temperature remains above 10oC was at least 112 days in 2005 and 94 days in 2006. In 2005, the season was longer than 112 days, but the sensor was not in place early enough to measure the date at which the soils temperature reached 10oC.  78 Figure A.1: Soil Temperature 0-10 cm Depth from May to October 2005 Figure A.2: Soil Temperature 0-10 cm Depth from May to October 2006   79 Low soil temperatures have a number of effects including reduced nodulation and NF by legumes (Rice et al. 1995; Sparrow 1988), slower germination, and slower growth (McKenzie et al. 1988). Soil temperatures at seeding are critical for germination, with more rapid germination in the 15-20oC range and most rapid growth of seedlings at 20-30oC  (McKenzie et al. 1988). In a study by McKenzie (1988) alfalfa growth rates were greatly reduced outside an air temperature range of 10-37oC. Root growth can also be limited at 10oC and soil temperatures below 15oC have been shown to completely inhibit or severely restrict nodulation (McKenzie et al. 1988).  Soil Moisture Soil water content varied between each location and throughout each site (Table A.4). The permanent wilting point for the silt loam soil at the research site was determined from a series of water retention curves to be 9-12% for 0-6 cm depth and 6- 9% for 6-12 cm depth (data not shown). Monitoring of permanent wilting points indicates the point at which plants can no longer draw water from the soil. At -15 bars a water deficit results and plant stress is exhibited by wilting (Brady and Weil 2002). Soil water was monitored using a Watchdog Time Domain Reflectrometry (TDR) sensor (Spectrum Technologies Inc).  Table A.4: Volumetric Moisture Content of Fields at the Yukon Government Research Farm 2005 and 2006   2005    2006  June July Aug Sept  June July Aug Sept Smooth bromegrass/alfalfa VMC (%) 21 23 24 25  20.5 21 20 13 Timothy/alfalfa VMC (%) 22 22 12 15  21 22.5 20 10.5 Alfalfa VMC (%) 16 23 9 11  12 10.5 12 6  Of particular concern is the frequency of occurrence of low water in the monoculture alfalfa plot (Table A.4). Frequent observations in the field noted wilting vegetation on only a couple of occasions - not as frequently as the TDR suggests. The difference between the observational notes and the TDR data can likely be attributed to the dry upper root zone. The TDR probe calculates one moisture level value for 0 - 20 cm depth. Since this value represents the average for the volume, it is not always representative of moisture content at greater depths which are accessible to the roots  80 (eg: soil water readings at 20 cm depth and 5 cm depth can differ substantially). Therefore, skewed averages resulting from dry upper soils would not account for moisture that is accessible deeper in the root zone.  81 Appendix B Detailed Methodology  Soil Compaction Soil compaction was measured at each site in the spring of 2005 to help describe the soil conditions. The soil compaction probe used was an analog meter from DICKEY- John®. Samples were taken from depths of 0-7.5 cm, 7.5-15 cm, and 15-30 cm.  Soil Bulk Density Soil bulk density was determined using the core method. A cylinder of known volume was inserted into the soil to extract a sample. The sample was dried at 105oC, and weighed and divided by the volume of the cylinder to give bulk density. Samples were taken at depths of 0-5 cm and 5-10 cm.  Soil and Air Temperature and Moisture Monitoring Small battery operated data loggers manufactured by Onset® Computer Corporation were used to record ambient temperature throughout the growing season at the cooperator trials. These loggers were connected to temperature sensors (TMCx- HD). Air temperature was recorded from approximately 2 m above ground on a steel bar using Stevenson Screens for solar shielding. This height was used in order to avoid stagnant air movement at ground level, where boundary layers tend to occur (Oke 1987). A stainless steel 102 mm probe (TMC6-HC) was inserted into the ground to monitor soil temperature from 0 to 10 cm. Both of these sensors are accurate to ±0.5°.  The air and soil monitoring equipment at the Yukon Government Research Farm was more complex than those mentioned above. Air temperatures were monitored from a forestry weather station on site and data was provided electronically. Air temperature, precipitation and wind were recorded. Soil moisture and temperature monitoring was a combination of Aquaflex temperature and moisture sensors (Streat Instruments Ltd) with data capture on a CR-10 datalogger (Campbell Scientific Corporation Inc). Aquaflex sensors are 3 m long flexible plastic tape that record soil temperature with ±0.5% accuracy and volumetric moisture content using time domain reflectrometry (TDR) with ±2% accuracy. The sensors were buried 15 cm into the ground. The CR-10 captured temperature and moisture hourly throughout the season and was programmed to  82 provide a minimum and maximum value for each day. The sensors were installed in early May and removed in September.  Moisture content was also measured with a portable TDR sensor and gravimetrically.  A portable moisture sensor Fieldscout TDR 300 (Spectrum Technologies Inc) was used at all the sites. This system used time-domain reflectrometry to provide volumetric moisture content from 0-20 cm with ±3.0% accuracy when EC < 2 dS/m. This unit proved to be the most useful, the only issue was that the probes would bend when used at one of the cooperator trials where the compaction was extremely high.  The TDR method works off the same principles as radar, where a signal is sent out from a probe and the speed of return provides information. In the case of TDR the speed at which the signal travels from one probe to another provides information about the water content of the soil. It is a non destructive test that uses electrical currents to determine the moisture value.  The gravimetric method involved weighing the soil samples, drying the samples and then reweighing to determine the change in mass from wet to dry. This method is more time consuming and is destructive but is considered the standard.  The TDR measurements were compared to those from the gravimetric method to ensure the accuracy of the TDR measurements. The results (Table A.5) indicate that the measurement of soil water content using the TDR method is relatively accurate, some caution should be taken though as the TDR moisture values were almost always greater than those achieved by the gravimetric method.  Table A.5: Gravimetric vs Time Domain Reflectrometry Volumetric Moisture Content Gravimetric VMC (%) 17.1 23.1 17.1 15.2 18.6 21.9 18.3 16.9 TDR VMC (%) 16.5 23.5 20 21 23 23 21.5 21.3   83 TDR measurements were taken randomly throughout each experimental unit at various intervals over the summer.  Agroclimatic capability ratings were assessed for each season to put the research years (2005 and 2006) in perspective to other years. The number of growing degree days (GDD) were calculated beginning the fifth consecutive day of the year with daily mean temperatures above 5oC, and terminating the day of the first killing frost (- 2.2oC) occurring after July 15th.  Soil Nutrient Monitoring Soil nutrient monitoring was carried out using the traditional soil sampling method with destructive testing and also using PRS™-probes from Western Ag Innovations. Both methods are described below.  Traditional soil samples Soil samples were extracted using an Oakfield open ended soil probe with a 30 cm maximum sample depth. Composite samples from 0-15 cm (made up of 10 samples) were randomly taken within each trial in the spring to determine any nutrient requirements and establish a baseline. This was followed by more comprehensive soil sampling within each experimental unit in the fall.  Samples were either analyzed for NO3, P, K and SO4 or complete nutrients. The complete test included NO3, P, K, SO4, Ca, Mg, Fe, Cu, Zn, B, Mn, Cl, pH, EC, OM, BS and TEC. NH4 was also analyzed in some cases. Samples were kept cold and shipped to Norwest Labs of Edmonton, Alberta for analysis (see Appendix I).  At the laboratory, samples were dried and ground for analysis. Nitrate and SO4 were extracted with a weak calcium chloride solution. Phosphorus and K were extracted with a modified Kelowna extract. Iron, Cu, Zn and Mn were extracted using DTPA. Boron was extracted with hot water. Chloride was extracted with deionized water. Analysis of the extractants was by colorimitry. pH and salinity were measured using a 2:1 water:soil mixture. Organic matter was computed using the loss on ignition method.   84 Fertilizer was applied as required based on the standard soil test recommendations of Norwest Labs. Phosphorus was already high in the soils. Nitrogen was applied based on experimentally derived values, only K, S and B were applied to the soil.  PRS™-Probes (see Appendix F for further information) PRS™-probes from Western Ag Innovations, Inc were used to monitor nutrient availability in the smooth bromegrass trial. The PRS™-probes act as ion sinks through the principles governing electrostatic attraction. The counter-ion used for the ion exchange membrane is Na for the cation and HCO3 for the anion (Western Ag Innovations Inc 2004). The probes arrived in a cooler and were kept cold and moist until inserted into the ground. A pair of probes (one cation and one anion) were placed within the root zone 2.5 cm from a row of alfalfa down to a depth of 15 cm at 4 different locations within each experimental unit. Probes were incubated for 48 hours during a period of irrigation. This was repeated four times throughout the summer, starting the beginning of June, followed by incubations in early July, August and September. The short duration in the ground was to minimize competition from ion sinks.  Vegetation Sampling and Analysis In all cases, vegetation samples were harvested as close to 10% bloom as possible. Harvest was late in the season at the Research Farm in 2005 to allow for alfalfa bloom. In the mixed stands at the Research Farm two subsamples from each experimental unit 1 m2 were clipped for collection. In the monoculture stand random rows were harvested. Each sample consisted of above ground biomass, plant counts, stage of growth, percent basal cover, heights, plant ID, and weed incidence. Plant samples were harvested, placed in paper bags, and dried for 48 hours at 40oC then weighed for “as fed” weight. One hundred gram subsamples from each sample were grouped together per experimental unit in order to save money on lab costs. These composite hay samples were sent to Norwest Labs in Lethbridge, Alberta for moisture and protein analysis. The LECO dry combustion method was used to determine total N; protein is determined by multiplying the total N by 6.25 (Norwest Labs 2004). Moisture was determined at 104oC.   85 Nodulation was assessed visually in each experimental unit. Comprehensive nodule assessments were carried out in the monoculture alfalfa stand using the scoring criteria outlined by Rice (Rice et al. 1977). Nodule observations were quantified by scoring nodule position, nodule numbers and nodule size (Table A.6). Position is assessed by the ratio of the number of nodules within 5 cm of the crown to the remaining nodules. These assigned numerical values for the criteria are based on previous descriptions of nodule characteristics and Rice’s experience.  Table A.6: Modified Nodule Ranking Criteria Characteristic Criteria Score Number 5-20/plant 3  >20/plant 2  1-5/plant 1  None 0 Position 60-100% 2 (% crown) 20-59% 1  0-19% 0 Size 3-10mm 1 (diameter) <3mm 0  >10mm 0 (Rice et al. 1977)  Nitrogen fixation was determined using the total plant nitrogen difference method (ND). With this method, the N derived from the soil is withdrawn from the total amount of plant N. Estimations of how much N that is derived from the soil is achieved by simultaneous cultivation of a non-fixing reference crop in the same field as the N fixing crop. This non-fixing reference crop is usually a non-nodulating legume genotype or a graminoid. The amount of fixed N is then assumed to equal the difference between N accumulated in the fixing crop and the reference crop (Ledgard and Steele 1992; Danso 1995). This method is relatively cheap and easy to perform. The problem is the basic assumption with the method, that the two crops absorb soil N with the same efficiency, may sometimes be incorrect.    86 Seeding and Inoculation Alfalfa was seeded standalone or into preexisting Carlton smooth bromegrass or Climax timothy forage stands. Alfalfa Medicago sativa cv Peace seed was purchased from Prairie Seeds in 2005 and Quality Seed West in 2006. Seeding equipment delivered a seeding rate of 10 kg/ha in the mixed stands. The monoculture stand was hand seeded at 15 kg/ha. Alfalfa seed was inoculated onsite prior to seeding with peat based TagTeam® (containing both NRG-34 and PB-50), N-Prove® (containing only NRG- 34), or wettable powder based fungal inoculant JumpStart® (containing PB-50). All inoculants were supplied by Philom Bios Inc. N-Prove® formulation had a titre of 3.14e9 rhizobia per gram. The commercially available TagTeam® had an internal titre of 6e8 rhizobia per gram and 2.4e7 Penicillium per gram. In order to provide equal amounts of rhizobia per treatment the application of TagTeam® was increased to equal the higher titre of the N-Prove® formula. The TagTeam® formula is available in a 490 g bag for 100 kg of alfalfa seed. N-Prove® formula for alfalfa was custom ordered. A 490 g bag was used for 100 kg of alfalfa. Jumpstart® formula is available in a wettable powder in 80g for 100kg of alfalfa or 400g for 500 kg of alfalfa.  Each formula was mixed with the required amount of seed and water (1.5 litres per peat bag or 2 litres for Jumpstart®) to wet the sticking agent in the peat or powder.  Alfalfa was seeded into existing grass stands with a no-till seeder from Truax Company, Inc, model FLX-812. This 2.4 meter wide double disc seeder was adequate for the circumstances. Its hydraulics and weight allowed the discs to penetrate through the thatch and vegetation and into the soil in one pass without harming the existing grass stand. The seeder passed over the grass, cutting a groove in the soil in which the seed and treatment (if included) were deposited.  All Truax drills are equipped with depth bands on all disc openers. The size on the unit used allowed for a 3.5 cm penetration of the blade. The double disc openers create a v groove in the soil surface for the seed to be dropped into. According to the Truax manual, seeds should drop into the seed slot about half the depth of the disc penetration, which is ideal for alfalfa (1-2.5 cm is recommended for alfalfa) (Truax Company 2004). Press wheels roll behind the discs and press the seed into the sides of the furrow. The machine is equipped with a small seed box capable of handling small 1-  87 2 mm alfalfa seed. The small seed box contains gears that rotate and deliver accurate amounts of seed to the discs. Prior to seeding the machine was calibrated. Calibration procedures were followed from the Truax manual. The drive wheel was turned 12.5 times. The seed collected was weighed and divided by two. This result equals the seeding rate in lbs/acre. The gears were adjusted by moving the exposed flute to deliver 10 kg/ha. Even though the small seed box was adjusted there was some seed damage and irregular seeding rates in the field.  88 Appendix C Calculations, Conversions and Equations  Conversion of 1 sq m samples to hectares [grams/1m2] to [kg/ha]  # * 10  Conversion of [grams/2m row] to [kg/ha] (There are 11 rows per 4m2) # * 11 / 4 * 10  Conversion of [as fed weight] to [dry matter weight] (kg/ha) asfed weight - (asfed weight * asfed moisture / 100)  Conversion to Plant Nitrogen Concentration (Conversion from Protein % to Nitrogen %) Protein % / 6.25  Calculating N yield (kg/ha) DM weight * Plant N Concentration  Calculating contribution of alfalfa to N yield in mixed hay (%) N yield Alfalfa / Total N yield * 100  Calculation of Nitrogen Use Efficiency Recovery Efficiency = (N uptake fertilized – N uptake unfertilized)/Nitrogen Applied = the amount of N taken up by the crop as a fraction of the amount applied  Total Plant Nitrogen Difference Method Biological Nitrogen Fixation = N accumulated N fixing crop – N accumulated non N fixing crop Subtract the inoculated value against the control value.   89 Appendix D Statistical Equations and Analysis of Variance Data  All data was analyzed using statistical models in SAS® 9.1 (SAS Institute). Data was compared using P=0.05 and the conservative Bonferroni multiple comparisons method. Bonferroni testing is a stepwise, sequentially rejective, method for hypothesis testing. Analysis of Variance (ANOVA) tables were generated for each model.  The model used for analysis of dry matter, N yield and N fixation at the Yukon Government Research Farm was: Y=Tr R(Tr) Sp Tr*Sp/ss3.  The experiments could not be analyzed based on subsamples since the subsamples were combined for composite testing at the lab to reduce costs associated with the N analysis.  The model used for analysis of the randomized complete block design used for the data accumulated at the cooperator farms was: Y= B Tr B*Tr.  Tr = Treatment = Inoculum R = Repetition B = Block Sp = Split = Fertilizer or Harvest  Analysis of Variance data from the Yukon Government Research Farm (Tables A.7 to A.9) and from the cooperator site Rafter ‘A’ Ranch (Table A.10) are presented below.  90  91  92  93  94 Appendix E Experimental Designs   At the Yukon Government Research Farm the fields were laid out in split plot completely randomized design with inoculant as the main effect and fertilizer or harvest on the split. Alfalfa was seeded 1) standalone or into preexisting 2) smooth bromegrass or 3) timothy forage stands. At the cooperator sites the fields were laid out in randomized complete block designs into an existing 4) smooth bromegrass stand. See maps below.  Figure A.3: Layout of 1) Monoculture Alfalfa   Eight plots 2x2 m with 1 m spacing were laid out in a split plot completely randomized design (Figure A.3) with the inoculation treatment using the TagTeam® formula against a control with no inoculant. Seed was hand sown in rows with 20 cm spacing.  Figure A.4: Layout of 2) Smooth Bromegrass and Alfalfa Mixed Forage Stand  95  Figure A.5: Layout of 3) Timothy and Alfalfa Mixed Forage Stand  For the mixed smooth bromegrass/alfalfa (Figure A.4) and timothy/alfalfa (Figure A.5) stands the treatments included: C > Uninoculated alfalfa with 0 or 25 kg/ha/yr N fertilizer; PB > Alfalfa inoculated with Jumpstart® PB-50 Penicillium bilaii and 0 or 25 kg/ha/yr N fertilizer; NP > Alfalfa inoculated with N-Prove® NRG-34 Ensifer meliloti and 0 or 25 kg/ha/yr N fertilizer; TT > A dual inoculation with TagTeam® (NRG-34 + PB-50) and 0 or 25 kg/ha/yr N fertilizer; Yield > No alfalfa with 0 or 170 kg/ha/yr N fertilizer. Figure A.6: Layout at Rafter ‘A’ Ranch  At the cooperator sites (Figure A.6) the trials were setup in randomized complete block designs with the same treatment as the previous layout with one addition: a treatment with 170 kg/ha N and alfalfa seeded: 170 N C > Alfalfa with 170 kg/ha N fertilizer.  96 Appendix F Understanding Nutrient Flux using Plant-Resin Simulator Probes  One of the objectives for the research forming this thesis was to monitor changes in soil nutrient supply and transfer in a subarctic mixed smooth bromegrass and alfalfa sward, under irrigated conditions, using Plant Resin Simulator (PRS™) Probes from Western Ag Innovations Inc. This investigation occurred over the 2005 growing season to determine if we could detect subtle changes over time in the plant available supply of P and N using PRSTM-Probes.  PRS™ resin membrane ion exchange probes have been used extensively to track changes in available nutrients. Although there is no direct correlation of the data collected from the PRS™-Probes that can be made to a standard soil test, the probes allow the monitoring of plant available supply of nutrients over time, for a predetermined timeframe and allow comparisons amongst different incubations.  Ion exchange probes have been in use for over 50 years to measure nutrient availability. They have been adopted to measure the amount of plant available nutrient ions in the soils and the rates at which they are released (Qian and Schoenau 2002). The use of ion exchange resins in membrane form has been widely adopted due to ease of use and the ability to measure the flux to a defined surface (Qian et al. 1992).  Methods PRS™-Probes were supplied by Western Ag Innovations Inc. Probes arrived in a cooler with ice packs and were promptly stored at 5oC until use. Probes were placed in the soil into slits 15 cm in depth at eight different locations within each experimental unit. The slits were made within three cm of the row of alfalfa that was interseeded into the smooth bromegrass. Four nutrient cation exchange probes and four nutrient anion exchange probes were placed in the slits.  PRS™-Probes were inserted in the ground for 48 hours during a period of irrigation. This was repeated at the beginning of each month throughout the summer. The short duration that the probes were in the ground was intended to minimize competition from ion sinks, but to be long enough to pick up the nutrient trends.   97 Alfalfa interseeded into the smooth bromegrass stand was inoculated onsite. Main plot treatments were: Control > Uninoculated alfalfa Jumpstart® > Alfalfa inoculated with Penicillium bilaii isolate PB-50 N-Prove® > Alfalfa inoculated with Ensifer meliloti isolate NRG-34 TagTeam® > A dual inoculation with NRG-34 + PB-50 Yield > Monoculture grass with no alfalfa seeded and 0 or 170 kg/ha/yr N fertilizer.  Each of the main plots was split, with half receiving 0 and half 25 kg N/ha as urea (46-0-0).  At the end of the allotted burial time-frame, the PRS™-Probes were removed and washed with deionized water to remove soil and plant material. They were then placed into sealable plastic bags and shipped to Western Ag Innovations Inc for testing.  In the lab the probes were placed in a dilute acid or salt solution to elute the absorbed ions into solution and measured analytically for nutrient concentration.  The Donnan exchange principle is based on counter ions Na and HCO3, which can be easily desorbed from the resin, thereby allowing nearby soil solution ions to be absorbed. Saturating the ion-exchange membrane with the same counter-ion ensures consistent ion-exchange chemistry across the entire PRS™-Probes’ ion-exchange membrane, thus increasing the measurement precision among treatments (Western Ag Innovations Inc 2004).  Ion supply rates were defined as the amount of ion absorbed per amount of ion exchange surface area, per time of direct burial in the ground. The resultant ion supply rate was therefore reported in mg/10cm2/48 hr.       98 Table A.11: PRS™-Probe Burial Dates and Irrigation and Moisture Levels in 2005  Irrigation over 48 hr burial (mm) Volumetric Moisture Content at start of burial (%) June 8-10 21 16 July 10-12 17 25 August 12-14  22 20 September 12-14 24 21  Throughout the growing season the plant height of the alfalfa was measured to provide an approximate growth curve.  Results  The alfalfa growth curve (Figure A.7) shows the steep shoot biomass gain from June 20th to July 15th. Figure A.7: Alfalfa Growth Curve in the Smooth Bromegrass Stand 2005    99 Seeded May 28th 20 days after planting = June 16th 40 days after planting = July 6th 60 days after planting = July 26th 80 days after planting = August 15th 100 days after planting = September 5th   The results from the incubations are graphed to assess the supply rate of the plant-available N and P over time. Maximum and minimum values are shown in the high and low bars. Figure A.8: Nitrate Levels in the Soil through the 2005 Season  100 Figure A.9: Phosphate Levels in the Soil through the 2005 Season  Discussion In previous studies the ion uptake by the PRS™-Probes was consistently related to plant uptake (Qian and Schoenau 2002; Western Ag Innovations Inc 2004). In the N fertilized samples shown in Figure A.2, the level of NO3 was high during the June incubation, which occurred 6 days after N (0-0-46) was applied to the fertilized treatments. In subsequent incubations the NO3 level dropped off and stayed low through the remainder of the season. This is indicative that there is a considerable amount of plant available nitrate in the soil immediately after fertilization and substantial plant uptake though June. This is surprising considering that the fertilizer used was a urea based product that requires microbial nitrification for the mobile NO3 to become accessible for plant uptake (Knight and Sparrow 1993). In the case of the subarctic environment with lower soil temperatures, urea release is thought to occur slowly. It is worth examining microbal acitivity in this context further as the subarctic soils may be more dynamic than previously thought.   101 In the unfertilized samples, the level of NO3 in the soil remains low through the entire season. No differences between the inocula is apparent in Figure A.8.  The P levels show a different trend, with high phosphate levels in June, a substantial drop in July, an increase in August and a drop in September. Phosphorus levels closely follow the plant growth curve.  Plant demand for P is greatest during the steep slope of the growth curve (in mid June to mid July) which is indicated on Figure A.9 by the drop in available P in the July sample. The plant available P level increased later in the season, likely because of increased mineralization of soil P from the microflora present within the soil releasing P solubilizing acids, and from the water solubilizing P.  There was no noticeable effect on P levels from the Jumpstart® (Penicillium bilaii isolate PB-50) treatment. This inoculum treatment was included to determine if there was extra benefit gained by the PB-50 acids solubilizing P which occurs up to 8 cm from the root (Kucey et al. 1989).  Conclusion PRS™-Probes are a useful method to investigate soil nutrient flux and there is reason to use this technology in northern environments. Nitrate supply rates were affected by the  low fertilizer level and P levels were affected by plant growth. It is very important that in cold soils and dry conditions the incubation times be longer than normal. In irrigated grass stands, as occurred in this research, the incubation times should be increased to at least 96 hours to be able to observe more subtle differences in the nutrient flux. Further study needs to be conducted on the contribution of the incubation time and soil moisture level to the ion flux in subarctic cold climate conditions.  Western Ag Innovations Inc. contact information: #3-411 Downey Road Saskatoon, SK S7N 4L8 Phone: (306) 978-0373  102 Appendix G Cooperator Farm Sites  Introduction Experiments on cooperator farms were designed to provide a comparison of dry matter (DM) and nitrogen (N) yield between high fertilizer rates of 170 kg/ha N historically applied to monoculture smooth bromegrass (Bromus inermis Leyss cv Carlton) stands, and a binary alfalfa/smooth bromegrass forage stand with low fertilizer. The hypothesis tested was that the seeding of inoculated alfalfa (Medicago sativa L. cv Peace) into an existing smooth bromegrass stand while limiting N fertilization would lead to greater N yield at harvest than a monoculture smooth bromegrass sward with a higher fertilizer rate.  Field experiments were conducted in the 2005 and 2006 growing season in the Takhini Valley, located west of Whitehorse, Yukon Territory. The experiments were set up within large cleared fields on two separate farms. Unfortunately, one of the seeded sites established poorly and was removed from the experiment midway through the first season. The remaining location, Rafter ‘A’ Ranch, was seeded in a large 40 hectare field on a gently north facing slope at 700 meters elevation. In the late 1990s this field was cleared of vegetation, which was mainly aspen, willow, grasses, bearberry, and sheperdia spp. In 2000, it was initially seeded to oats and undersown with a forage mix, containing smooth bromegrass.  The producer raises cattle and boards horses on the property. The animals are kept off the hay fields from the spring until after harvest in July. On assessment in May 2005, the forage stand was dominated by smooth bromegrass, with some other species present: Jacob’s Ladder (Polemonium occidentale Greene), Yarrow (Achillea millefolium L.), Rose (Rosa L.), Narrow-leaved Hawksbeard (Crepis tectorum L.), Foxtail Barley (Hordeum jubatum L.), and Kentucky Bluegrass (Poa pratensis L.).  The plow layer extends down 25 cm from the surface and consists of brown clay loam to silty clay loam over grey calcareous silty clay parent material. Silty clay to clay loam Orthic Brunisols at the site were developed from fine textured glacio-lacustrine deposits. The soil association is Champagne, which is described as moderate to well  103 drained and is moderately permeable with good moisture holding capacity (Day 1962). The Champagne soil series covers 44,000 ha and is the most extensively used agricultural soil in the Yukon. Average soil pH for Rafter ‘A’ Ranch is 7.7, while soil organic matter is 3% by weight. The organic matter level is 1% higher than the baseline found in most south central Yukon conditions, likely due to the number of years this field has been in permanent grass cover. Also, the north aspect of the slope allows for the retention of moisture over the summer promoting biomass growth and slowing decomposition, providing a better regime for the accumulation of organic matter. Electrical conductivity was < 1 dS/m indicating that there are no salinity concerns. As is the case in most Yukon soils, the soils at this location have limited available N. Background NO3 levels were always below 4 ppm.  Rafter ‘A’ Ranch had a much warmer growing season in 2005 in comparison to the Yukon Government Research Farm located nearer to Whitehorse. In 2006, all experimental sites had similar seasons.  Materials and Methods This trial was laid out in a randomized complete block design. Each block was split into six experimental units. Each experimental unit was 2.4 x 20 m. Four blocks were laid out in the 2005 season (blocks 1 - 4) and three additional blocks were added in the 2006 season (blocks 5 - 7). Alfalfa was seeded into the pre-existing irrigated smooth bromegrass hay stand using a Truax FLX-821 no-till seeder. The seeder passed over the grass, cutting a groove in the soil in which the seed and treatment (if included) were deposited.  Alfalfa was inoculated onsite with peat based TagTeam®, N-Prove®, or JumpStart® inoculant. All inoculants were supplied by Philom Bios.  Treatments included: 25 N Control > Uninoculated alfalfa with 25 kg/ha/yr N fertilizer; 25 N Jumpstart® > Alfalfa inoculated with PB-50 penicillium bilaii and 25 kg/ha/yr N fertilizer; 25 N N-Prove® > Alfalfa inoculated with N-Prove® NRG-34 rhizobia and 25 kg/ha/yr N fertilizer; 25 N TagTeam® > A dual inoculation with NRG-34 + PB-50 and 25 kg/ha/yr N fertilizer; 170 N Control > Uninoculated alfalfa with 170 kg/ha/yr fertilizer; 170 Yield > Monoculture grass with no alfalfa seeded and 0 or 170 kg/ha/yr N fertilizer.  104 Seeding occurred May 10, 2005 and May 18, 2006 and was seeded at 10 kg/ha. The site was irrigated approximately with 150 mm in 2005 and 130 mm in 2006 from May to late June.  Composite soil tests were taken in each block in the spring. Based on recommendations from Norwest Labs, fertilizers were applied to bring the soil to ideal nutrient levels for a grass/legume mixed forage stand. Nitrogen (46-0-0) was applied based on experiment driven values along with 47 kg/ha P (0-45-0), 34 kg/ha K (0-0-62), 17 kg/ha S (0-0-0-90), and 2 kg/ha B (as Borax). All fertilizers were broadcast applied in the spring.  Harvest occurred July 6, 2005 and July 10, 2006. The smooth bromegrass was sampled at the early boot stage in both years. Each treatment was divided into a grid and a number assigned to each grid point. Within each block, four randomly selected samples were taken from each treatment with 0.5 x 0.5 m2 sampling squares. Each sample consisted of above ground biomass, plant counts, heights, stages of growth, and weed incidence.  Results The majority of the alfalfa plants were in the first trifoliate stage in early June 2005, three weeks after planting. Alfalfa growth was best in areas with sparse grass cover. Alfalfa production was poorest in the 170 kg/ha N control treatment and the best production occurred in the N-Prove® and TagTeam® treatments.  Results were analyzed based on shoot biomass dry matter and N yields (Tables A.12 to A.14). Alfalfa dry matter yields achieved up to 118 kg/ha and 1.5 kg/ha of N in the establishment year. Nitrogen contribution to the stand did not exceed 5%. In the second year stand the alfalfa DM and N yields increased over the establishment year, with up to 246 kg/ha of DM and 7.8 kg/ha N.  Regrowth was monitored into the fall in 2005. Alfalfa regrowth was best with the TagTeam® and N-Prove® inoculated treatments. The height and colour of these treatments was generally as good as or better than the regrowth of the monoculture smooth bromegrass fertilized with 170 kg/ha N. Unfortunately, in September 2005 the  105 horses that were boarded on the property were let onto the field and within a short time had discovered the alfalfa. The horses preferentially grazed the alfalfa rows. As discovered in chapter 2, the result of late season grazing likely resulted in reduced yields in year two.  In the fall of 2005, composite soil samples were taken for each experimental unit and analyzed for complete nutrients or for N, P, K, and S. Nitrate levels were near 0 ppm for all samples. There was very limited variability in the other tested nutrient levels across the site.  Soil moisture TDR measurements were taken within each block. Soil moisture was maintained at an average of 43% throughout the peak growing season in 2005 and 2006. The undulating terrain at the site provided different moisture characteristics. Block four, in a low lying portion of the field, had 10% greater moisture throughout the season.    106  107 Discussion Alfalfa contribution in terms of N reached 17% of the mixed smooth bromegrass/alfalfa stand with 25 kg/ha N fertilizer by the second year in the TagTeam® treatment. Even with the legume incorporated into the stand the total biomass of the mixed stands achieved at best only half of the DM in the monoculture smooth bromegrass stand fertilized at 170 kg/ha N. In the monoculture treatment yields ranged from 6.7 to 8.7 t/ha.  The total N yields of the mixed stands were 1/3 of the N provided in the 170 kg/ha N fertilized treatments.  It is interesting to note that the total DM yields of the 170 N Control treatments are all lower than those of the 170 N Yield treatments. This might be an artifact of the no-till drill cutting through the soil and injuring the grass roots, slightly reducing yields. In the 170 N Control treatments DM yields ranged from 6.1 to 7.5 t/ha, approximately 8- 16% lower than yields for the 170 N Yield treatment.  Sample sizes were 0.25 square meters for the cooperator trials. This was a realistic sample size given the amount of labour involved in the cutting and measuring of the vegetation; however, it did not allow for capture of the field variability, as is evident by the standard deviations in the figures. One square metre, as was used at the Yukon Government Research Farm for the other experiments, would have been more accurate.  Working within fields on farm brought with it the challenge of ensuring the goals of the research and the field layout were clearly communicated to all who worked in the area where the research was being conducted. However, it also came with the benefit of being able to immediately share results with interested farmers in the area.  108 Appendix H Growing Alfalfa in the Yukon Yukon Government Extension Bulletin 2008  Alfalfa (Medicago sativa) is a drought tolerant, high protein, high yielding, perennial legume that can be a stand alone forage crop or included in a mixed forage stand. As a legume, alfalfa is capable of providing its own Nitrogen from the air. Nitrogen is the most limiting nutrient to crop growth in almost all Yukon crops; therefore, incorporating this crop into a seed mix or in rotation allows for much needed Nitrogen in the farm soil/crop system. Although alfalfa is a hardy plant, the Yukon’s climatic conditions are on the edge of this plant’s tolerance. Below are some points to keep in mind if you are thinking of growing and maintaining an alfalfa stand in the Yukon.  Variety Selection Variety selection is important to gain maximum yields and maintain stand longevity. There are 240 registered alfalfa cultivars through the US National Alfalfa and Forage Alliance (National Alfalfa and Forage Alliance 2007) with a range of characteristics including 11 fall dormancy ratings, 6 winter survival ratings, and 5 disease resistance ratings. The Canadian Food Inspection Agency maintains the registration list for available Canadian cultivars. The CFIA list includes 163 cultivars (Canadian Food Inspection Agency 2007). There are a number of hardy varieties for Yukon conditions, among these are the Peace and Anik varieties. Peace is often still used as the check variety in research trials. As well, Peace has historically produced greater biomass in Yukon when compared to other varieties. There are new cultivars being touted by seed companies as higher producers and with greater winter hardiness that may work in Yukon conditions. Select varieties that have good winterhardiness, high fall dormancy ratings and resistance to diseases - especially root rots and bacterial wilts.  Inoculation In order for alfalfa to successfully accumulate N from the air, the plants have to form a bond with bacteria known as rhizobia. These bacteria are attached to the seed and after seed germination infect the root of the alfalfa. Inoculant comes as pretreated seed, as a peat based formula or as a granular formula. Pretreated seed is the most  109 commonly used method of inoculation today. Rice et al. (2001) compared peat-based and preinoculated seed and found that rhizobium numbers are usually greater on freshly inoculated seeds than on pre-inoculated seeds, so it can be extrapolated that onsite seed inoculation provides greater assurance that the roots will achieve adequate nodulation. These are living organisms and it is important that these bacteria be kept in cool, dry, dark conditions prior to seeding. There are a number of different rhizobia that are used to inoculate alfalfa. Most frequently used in Canada is the pre-inoculated seed with Nitragin, Inc cultures. Another good option that has been tested in the Yukon is a peat based inoculum from Philom Bios called TagTeam®.  To see if the alfalfa is in fact accumulating nitrogen, check the root nodules for pink colouring using a fingernail to split the nodule in half. Pink colouring in the nodules indicates the presence of leghemoglobin, which is similar to our blood hemoglobin, which binds to the oxygen in the root nodules to allow for the oxygen sensitive dinitrogenase enzyme to fix atmospheric N (Dakora 2003). This colour shows that the rhizobia and alfalfa are creating plant available N.  Overall, the inoculation and subsequent infection to form nodules is critical. In their absence plant protein content and yield will be low (Rice and Olsen 1992; Sparrow 1988).  Soil Conditions Proper nutrient levels and ratios are key to successful production and stand longevity. Other soil conditions to consider for field selection include pH, salinity and compaction.  Fertilization is based on the replenishment of nutrients extracted by the crop and takes into consideration the immobilization, leaching, or volatilization of some nutrients that are lost from the soil or otherwise unavailable for plant uptake. The single most important controllable environmental factor affecting N fixation under field conditions is soil N. Adding fertilizer N to legumes decreases the fraction of plant N derived from NF (Vance et al. 1988). Although alfalfa does not need fertilizer N, many other nutrient additions are required in order to maintain balanced soil/plant nutrition. Alfalfa  110 continuously depletes soil nutrients. An 11 t/ha alfalfa crop removes 69-85 kg/ha of P2O5, 300-370 kg/ha K2O and 30-37 kg/ha S.  Proper fertilization helps in stand establishment, overwintering, persistence, promotion of early growth, increasing yield and quality, and improving disease resistance. Takhini Valley soils are only weakly weathered and therefore contain most required micro nutrients in sufficient quantities, but some key macro nutrients are in low supply. Alfalfa production removes a range of nutrients from the root zone which must be replenished. Nutrient levels are easily monitored with conventional soil tests and amendments can be applied where needed.  As with all other crops, pH affects the growth of alfalfa. Alfalfa flourishes in pH of 6.5-8; in acidic conditions (pH < 6.0) its production is substantially reduced. Slightly alkaline soils are found throughout the south central Yukon, providing ideal pH for alfalfa establishment.  A study by Panciera and Sparrow (1995) showed that in acidic soils in the Delta Junction area of Alaska the DM yields were 5 times less on soils with pH of 5.3 compared to 6.5-7. Field trials in Alaska compared alfalfa production in neutral (pH 7.2) and acidic (pH 5.4) soils and found yields reduced by almost 50%  in acidic conditions (Sparrow et al. 1993). Rice (1977) studied alfalfa growth in a range of pH at Beaverlodge, Alberta, and found a negative trend in nodulation scores and alfalfa yield as soil pH decreased from 6.0 – 5.0 (Rice et al. 1977).  Saline conditions exist in many places in the Yukon, typically in depressional sites adjacent to slopes, where the accumulation of salts occurs near the surface from the evaporation of water. Yield losses can be expected if saturated electrical conductivity levels are above 2 dS/m in the upper soil profile.  Alfalfa is most productive in deep, uncompacted, well-drained soils, which allow for deep root development. Under severe compaction alfalfa will have trouble becoming established.  It is important to soil test before planting in order to determine which nutrients are needed and to check the pH and salinity level of the soil.   111 Seeding Alfalfa is a small seeded crop with a variable seeding depth dependent on the soil texture. Heavy soils should be seeded 0.6-1.25 cm; in light soils, seed can be placed 1.25-3.8 cm into the soil. Seed should be placed near the maximum depth to access greater soil moisture, yet placing the seed deeper risks that the hypocotyl will not reach the soil surface. Seeding rates should be around 10 kg/ha. The small initial size of the alfalfa seedling reduces its competitive ability (Fick et al. 1988), so earlier seeding and germination is best. In dryland conditions spring rainfall distribution for germination and cold soils at planting can present problems for establishment, this has been shown in Alaska (Panciera and Sparrow 1995).  Irrigation Although alfalfa is drought tolerant, the Yukon’s semi-arid status results in precipitation levels that are below the minimum crop water demand for a good yielding crop. Irrigation will provide higher yields in this environment. In addition, a lack of precipitation at seeding can result in poor stand establishment.  Alfalfa requires replenishment of the evapotranspiration (ET). Factors that affect ET include solar radiation, temperature, wind speed, humidity and also stage of plant growth. ET in the south central Yukon ranged from 344 - 405 mm in 2005 and 2006 with precipitation providing 40% of the required water. The remaining gap between ET and precipitation is closed through irrigation. Non-irrigated crops need to be planted as early as possible in the spring to take advantage of existing soil moisture (Panciera and Sparrow 1995). For non-irrigated crops, seed should be planted as early as possible in the spring to take advantage of existing soil moisture.  It is important to maintain moisture levels to minimize plant stress, which ultimately affects DM yield and long term stand survival. On the other hand, excess soil water is also a concern as it decreases the growth of seedlings and promotes the development of fungal diseases.  Winter Survival Winterkill is the term used to describe the failure of an overwintering crop to survive low temperatures and concomitant cold injury (Leep et al. 2001). A plant's  112 capability to overwinter is based on the ability to go dormant, and to survive stresses through the winter including frost, frost heave, ice formation, low temperatures, and fungal and bacterial infection (Bertrand and Castonguay 2003). Snow depth is critical for insulating alfalfa crowns and preventing winterkill. Where possible, put up wind breaks perpendicular to the dominant winter wind direction to capture snow in field.  Alfalfa requires time in the fall to build up the root energy reserves to survive the long, cold winters in the Yukon. For best results, harvest should occur in mid summer, and wait to graze the regrowth after freeze up, around early November.  Diseases also cause winter death of plants. Yukon’s cold soils promote the growth of fungal diseases. Three diseases to watch for in field are verticillium wilt (Verticillium albo-atrum), Brown Root Rot (Phoma sclerotioides), and winter molds (Sclerotinia spp).  Symptoms of Verticillium Wilt – leaves become yellow, often in a v-shaped or one-sided pattern. Plants wilt and xylem tissue of taproot becomes brown. Stem remains green for awhile after all the leaves on a shoot are killed. Verticillium wilt is primarily found on irrigated fields (Stuteville and Erwin 1990).  Symptoms of Brown Root Rot – the taproot is destroyed around the plow pan level. Infected plants appear weak and are easily pulled up by hand. Symptoms include dark, necrotic lesions of the lateral and taproots, which are brown with a blackish border (McKenzie and Davidson 1975). Young lesions are generally brown and circular with dark perimeters. Maturing lesions generally expand asymmetrically and eventually girdle the root. Rhizobium nodules can also be infected and rotted. The fungus grows at subzero temperatures in the range of -7oC to 27oC (Stuteville and Erwin 1990).  Symptoms of Winter Molds (also known as snow molds) – these are most damaging to seedling stands, which are seeded in the late summer, a practice not recommended in the Yukon. Produces a cotton weblike growth on the stems and crowns of infected plants (Stuteville and Erwin 1990). Individual stems in patchy areas wilt.   113 The alfalfa stand will quite often rebound from the disease of concern when conditions improve but reduced yields are likely. These diseases can be somewhat managed by control plowing, spring planting, disinfecting equipment, maintaining proper soil fertility, and using alfalfa cultivars that have resistance to the disease. A considerable amount of time has been used in the breeding programs for disease resistance. Check the disease ratings for each variety and use disease resistant cultivars if available and check to see if that cultivar is suitable for this northern climate.  Weed Control Weed control can become a concern in a poorly established stand, so the first defence against weeds is to maintain a healthy, vigorous monoculture or mixed stand. In circumstances with high populations of broad leaf weeds, there are a few options for weed management including chemical applications and mechanical weeding.  Synthetic auxins are used for a range of broad leaf weed species. 2,4-DB (or Embutox 625), is specific for application on alfalfa crops. 2,4-DB affects protein synthesis and normal cell division, disrupting plant cell growth in newly forming stems and leaves leading to malformed growth and tumors (Alberta Agriculture 2007). The product should be used early in the year between the first trifoliate and the fourth trifoliate. Read the product label for specific instructions and mixing instructions.  Other options to minimize competition from weeds include mechanical removal of weeds by handpicking, tilling, cutting or flaming. Many broad leaf weeds are annual or biennial, therefore requiring seed production for propagation. If no seeds are set, the weed population will slowly diminish in the stand over time.  Planting into Mixed Stands Be wary of mixing brome and alfalfa in the same stand if you are planning on adding N fertilizer (>25 kg/ha), or allowing free range grazing. There is little point in including the alfalfa in the mix as it will be outcompeted for resources very quickly and will be reduced in the stand over a couple of years.   114 Crop Rotation There are a number of options for cropping alfalfa including seeding into an existing grass stand, seeding as part of a forage mix, seeding with a companion crop or seeding as a stand-alone crop. When seeding as a stand-alone crop, alfalfa is a suitable crop to use in rotation with other field crops to increase soil available N for the subsequent crops. It is important to select crops that are not susceptible to the same diseases. Oilseeds, cereals, and some vegetable crops are good options.  Harvesting and Grazing Management Harvesting and grazing management is one of the keys to successful alfalfa stand establishment in the North. In order to grow alfalfa in the Yukon, farmers will need to harvest early, by mid July at the latest, and must wait until after freeze up, around early November to graze the regrowth. Late fall grazing or cutting prior to freeze up, cuts the leaves that should be synthesizing substances that are translocated to the crowns and roots (McKenzie et al. 1988). Cutting during this period interferes with the accumulation of food reserves because new growth is produced at the expense of winter reserves.  Cracking alfalfa stems at harvest is good practice to reduce the drying time in the field. The faster the hay dries and the more uniformly it dries, the better the hay quality and less leaf loss. Reconditioner units with rollers crack alfalfa stems, drying them more quickly. In years where excessive moisture is a problem at harvest large round bales can be produced and wrapped in plastic for either haylage or silage.  Rotational grazing is a good option for alfalfa in a pasture, as opposed to straight grazing, which allows the preferential grazing of the alfalfa from the stand ahead of the grass. Straight grazing pressure will result in reduced alfalfa plants in the stand over time (Papadopoulos et al. 2005). Animals in a rotational grazing are removed after they have grazed the area thoroughly, allowing the pasture to regrow, in contrast to straight grazing where the animals have free access to all the plants and can preferentially graze legumes. Harvesting has a rapid negative effect on NF with declines of 5 - 30% reported (Vance et al. 1988). It is important to note that frequent cutting can increase protein content, but is also associated with slow regrowth, plant death, weed invasion and drastic yield reductions (Vance et al. 1988).  115 Feeding Alfalfa is a high protein, high calcium, high energy forage that can be used as feed for horses, dairy goats, dairy cattle, beef cattle, and game farm animals, and is also sold through pet stores for rabbits and guinea pigs. A high protein source, with upward of 20% protein levels, feeding alfalfa can result in overheating and colic in inactive horses. Care must be taken in feeding this ration to sedentary animals. Often veterinarians will recommend that only a small percentage of the hay feed mix be alfalfa (Ewing 1997). As when changing any feed ration, farmers must be watchful for changes in the animal's behavior and, if there are any concerns, consult a veterinarian.  If you have any questions about alfalfa production in the Yukon please consult the Yukon Agriculture Branch at 867-667-5838.   116 Appendix I Soil Nutrient Analyses  Soil samples were extracted using an Oakfield open ended soil probe with a 30 cm maximum sample depth. Composite samples from 0-15 cm (made up of 10 samples) were randomly taken within each trial in the spring of 2005 to determine any nutrient requirements and establish a baseline. This was followed by more comprehensive soil sampling within each experimental unit in October of 2005 and 2006.  Samples were either analyzed for NO3, P, K and SO4 or complete nutrients. The complete test included NO3, P, K, SO4, Ca, Mg, Fe, Cu, Zn, B, Mn, Cl, pH, EC, OM, BS and TEC. NH4 was also analyzed in some cases. Samples were kept cold and shipped to Norwest Labs of Edmonton, Alberta for analysis.  Tables A.15 to A.19 present the complete soil data from the sampling at the Yukon Government Research Farm.  117   118   119   120   121   122 References  Alberta Agriculture. 2007. Crop Protection. Alberta Government, Edmonton, Alberta, Canada. 539 pp. Bertrand, A. and Castonguay, Y. 2003. Plant adaptations to overwintering stresses and implications of climate change. Can. J. Bot. 81: 1145-1152. Brady, N. C. and Weil, R. R. 2002. The nature and properties of soils. Thirteenth ed. Pearson Education, Patparganji, Delhi, India. 960 pp.  Canadian Food Inspection Agency. 2007. List of varieties which are registered in Canada. Government of Canada, Ottawa, Ontario, Canada. 35 pp.  Dakora, F. D. 2003. Defining new roles for plant and rhizobial molecules in sole and mixed plant cultures involving symbiotic legumes. New Phytol. 158(1): 39-49.  Danso, S. K. A. 1995. Assessment of biological nitrogen fixation. Fert. Res. 42: 33-41.  Day, J. H. 1962. Reconnaissance soil survey of the Takhini and Dezadeash Valleys in the Yukon Territory. Research Branch Canada Department of Agriculture, Ottawa, Ontario, Canada. 53714-2: 1-78.  Environment Canada. 2008. Canadian climate normals 1971-2000. Government of Canada, Ottawa, Ontario, Canada. www.weatheroffice.gc.ca. Ewing, R. A. 1997. Beyond the hay days. 1st ed. PixyJack Press, LLC, Colorado, USA. 128 pp. Fick, G. W., Holt, D. A., Lugg, D. G. 1988. Environmental physiology and crop growth. Pages 163-194 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. First ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA. Heichel, G. H. and Henjum, K. I. 1991. Dinitrogen fixation, nitrogen transfer, and productivity of forage legume-grass communities. Crop Sci. 31: 202-208. Hill, T., Beckman, D., Ball, M., Smith, P., Whelan, V. 2002. Yukon agriculture state of the industry, 2000-2001. Yukon Government, Whitehorse, Yukon, Canada. 56 pp. Knight, C. W. and Sparrow, S. D. 1993. Urea nitrogen budget for a subarctic agricultural soil. Soil Sci. Soc. Am. J. 57: 1138-1144. Kucey, R. M. N., Janzen, H. H., Leggett, M. E. 1989. Microbially mediated increases in plant-available phosphorus. Adv. Agron. 42: 199-228. Lanyon, L. E. and Griffith, W. K. 1988. Nutrition and fertilizer use. Pages 333-372 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. 1st ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA.  123 Ledgard, S. F. and Steele, K. W. 1992. Biological nitrogen-fixation in mixed legume grass pastures. Plant Soil 141(1-2): 137-153. Leep, R. H., Andresen, J. A., Jeranyama, P. 2001. Fall dormancy and snow depth effects on winterkill of alfalfa. Agron. J. 93(5): 1142-1148. McKenzie, J. S. and Davidson, J. G. N. 1975. Prevalence of alfalfa crown and root diseases in the Peace River region of Alberta and British Columbia. Canadian Plant Disease Survey 55: 121-125. McKenzie, J. S., Paquin, R., Duke, S. H. 1988. Cold and heat tolerance. Pages 259- 302 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. 1st ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA. National Alfalfa and Forage Alliance. 2007. Winter survival, fall dormancy and pest resistance ratings for alfalfa varieties. National Alfalfa and Forage Alliance. Kennewick, Washington, USA. 8 pp. Norwest Labs. 2004. Schedule of lab testing. Edmonton, Alberta, Canada. 25 pp. Oke, T. R. 1987. Boundary layer climates. 2nd ed. Taylor & Francis, Great Britain. 435 pp. Panciera, M. T. and Sparrow, S. D. 1995. Effects of nitrogen fertilizer on dry matter and nitrogen yields of herbaceous legumes in interior Alaska. Can. J. Plant Sci. 75: 129- 134. Papadopoulos, Y. A., McKenzie, D. B., McRae, K. B., Clark, E. A., Charmley, E. 2005. Evaluating the performance of alfalfa cultivars in rotationally grazed pastures. Can. J. Plant Sci. 85(1): 147-150. Qian, P., Schoenau, J. J., Huang, W. Z. 1992. Use of ion exchange membranes in routine soil testing. Commun. Soil Sci. Plant Anal. 23: 1791-1804. Qian, P. and Schoenau, J. J. 2002. Practical applications of ion exchange resins in agricultural and environmental soil research. Can. J. Soil Sci. 82(1): 9-21. Rice, W. A., Penney, D. C., Nyborg, M. 1977. Effects of soil acidity on rhizobia numbers, nodulation and nitrogen-fixation by alfalfa and red-clover. Can. J. Soil Sci. 57(2): 197-203. Rice, W. A. and Olsen, P. E. 1992. Effects of inoculation method and size of Rhizobium meliloti population in the soil on nodulation of alfalfa. Can J. Soil. Sci. 72(1): 57-67. Rice, W. A., Olsen, P. E., Collins, M. M. 1995. Symbiotic effectiveness of Rhizobium meliloti at low root temperatures. Plant Soil 170: 351-358. Rice, W. A., Olsen, P. E., Lupwayi, N. Z., Clayton, G. W. 2001. Field comparison of pre-inoculated alfalfa seed and traditional seed inoculation with inoculant prepared in sterile or non-sterile peat. Commun. Soil Sci. Plant Anal. 32(13-14): 2091-2107.  124 Rostad, H. P. W., Kozak, L. M., Acton, D. F. 1977. Soil survey and land evaluation of the Yukon Territory. Saskatchewan Institute of Pedology, Saskatoon, Saskatchewan, Canada. 495 pp. Smith, C. A. S. 1990. Nature of the cryic thermal regime of agricultural soils in the Yukon Territory, Canada. In Proceedings International Symposium on Frozen Soil Impacts on Agricultural, Range and Forest Lands 90-1: 11-20. Sparrow, S. D. 1988. Inoculation of alfalfa in Alaska. Agroborealis 20(1): 38-40. Sparrow, S. D., Cochran, V. L., Sparrow, E. B. 1993. Herbage yield and nitrogen accumulation by seven legume crops on acid and neutral soils in a subarctic environment. Can. J. Plant Sci. 73: 1037-1045. Stuteville, D. L. and Erwin, D. C., eds. 1990. Compendium of alfalfa diseases. 2nd ed. The American Phytopathological Society, St. Paul, Minnesota, USA. 84 pp. Tarnocai, C., Smith, C. A. S. and Beckman, D. 1988. Agriculture potential and climate change in Yukon. Third meeting on northern climate, Whitehorse, Yukon. Environment Canada, Downsview, Ontario, Canada. Truax Company. 2004. FLEXII drills operator's manual. Truax Company, New Hope, Minnesota, USA. 100 pp. Undersander, D., Becker, R., Cosgrove, D., Cullen, E., Doll, J., Grau, C., Kelling, K., Rice, M. E., Schmitt, M., Sceaffer, C., Shewmaker, G., Sulc, M. 2004. Alfalfa management guide. ASA-CSSA-SSSA, Madison, Wisconsin, USA. 58 pp. Vance, C. P., Heichel, G. H., Phillips, D. A. 1988. Nodulation and symbiotic dinitrogen fixation. Pages 229-258 in: Hanson, A. A., Barnes, D. K.,Hill, J. R. R., eds. Alfalfa and alfalfa improvement. 1st ed. ASA-CSSA-SSSA, Madison, Wisconsin, USA. Western Ag Innovations. 2004. PRS-probe operations manual. Western Ag Innovations Inc., Saskatoon, Saskatchewan, Canada. 59 pp.    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0066717/manifest

Comment

Related Items