Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development of health indicators for rough fescue grasslands in the southern interior of British Columbia Lamagna, Sarah Frances 2008

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

Item Metadata

Download

Media
24-ubc_2008_fall_lamagna_sarah_frances.pdf [ 2.52MB ]
Metadata
JSON: 24-1.0066928.json
JSON-LD: 24-1.0066928-ld.json
RDF/XML (Pretty): 24-1.0066928-rdf.xml
RDF/JSON: 24-1.0066928-rdf.json
Turtle: 24-1.0066928-turtle.txt
N-Triples: 24-1.0066928-rdf-ntriples.txt
Original Record: 24-1.0066928-source.json
Full Text
24-1.0066928-fulltext.txt
Citation
24-1.0066928.ris

Full Text

DEVELOPMENT OF HEALTH INDICATORS FOR ROUGH FESCUE GRASSLANDS IN THE SOUTHERN INTERIOR OF BRITISH COLUMBIA  by  Sarah Frances Lamagna B.Sc. State University of New York College of Environmental Science and Forestry, Syracuse, NY, 2006 —  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 August 2008 © Sarah Frances Lamagna, 2008  Abstract Grasslands throughout the world including those in British Columbia have been severely reduced and altered by agricultural production and inappropriate livestock grazing practices. Ongoing degradation of rangelands is a worldwide problem, currently affecting about 680 million hectares of rangelands. Studies on development and application of criteria and indicators for forests and grasslands are often lacking, or have been done on a limited number of sites with relatively narrow ranges of climate and soil type. This study aims to (i) quantif’ the relationships among soil/vegetation properties known to be affected by grazing to easily-assessed indicators, used in the existing health assessment systems, that do not require laboratory analyses or time consuming measurement, and (ii) to evaluate impacts of grazing on soil aggregate stability on the rough fescue grasslands of the southern interior of British Columbia. During the growing seasons of 2006 and 2007, soil and vegetation properties were measured on nine open grassland sites with a potential natural plant community dominated by rough fescue (Festuca campestris Rydb.) in the southern interior of British Columbia. Each site had at least one area excluded from grazing and all units were classified into different seral stages according to the amount of rough fescue present on the land. Rough fescue cover was found to be a useful indicator of the presence of functioning recovery mechanisms. Percent exposed mineral soil was found to be a sensitive indicator of the degree of soil stability and watershed function, as well as an indicator of the integrity of nutrient cycles and energy flows in rough fescue grasslands. Percent Junegrass cover was not as sensitive an indicator as percent exposed mineral soil, but has general overall strength with many health measures.. Only the 1-2 mm aggregate size class was closely related to most soil and vegetation properties, showing that it is more sensitive than the other aggregate stability parameters to soil and vegetation properties. The results from this study can help rangeland managers and ranchers in determining the rangeland health in their area as well as help researchers understand that only a certain number of parameters need to be assessed.  11  Table of Contents Abstract . .  .........  ii  Table of Contents  iii  ist of Tables 4 E.  vi  Iist of Figures. 4ckno4vledgeI3ents  xii  Co—i.uthorship Stateinent  xiii  1. General Introduction 1.1. North American Grasslands  1 3  1.1.1. Rough Fescue Prairie  4  1.1.2. Palouse Prairie  5  1.2. Impacts of Grazing on Plant Communities  5  1.2.1 Succession of British Columbia Grasslands  6  1.3. Impacts of Grazing on Soil Properties 1.3.1. Soil Physical Properties  7 7  1.3.1.1. Soil Compaction  8  1.3.1.2. Soil Structure  8  1.3.1.3. Water Infiltration Rate and Soil Water Content  9  1.3.2. Soil Chemical Properties  10  1.3.2.1. Soil Organic Matter  11  1.3.2.2. Soil Polysaccharides  13  1.4 Rangeland Health Assessment Methods  14  1.5 Summary of General Introduction  15  1.6 Study Objectives  17  1.7 References  18  2. Development of Health Indicators for Rough Fescue Grasslands of British Columbia 2.1 Introduction  22 22  111  2.2 Materials and Methods  .  24  2.2.1 Site Description  24  2.2.2 Sampling and Analyses  24  2.2.2.1 Soil Sampling  24  2.2.2.2 Plant Sampling  26  2.2.2.3. Grazing  27  2.2.3 Statistical Analysis  28  2.3 Results and Discussion  28  2.3.1 Principle Criteria no. 1 Degree of Soil Stability and Watershed Function  28  2.3.2 Principle Criteria no. 2 Integrity of Nutrient Cycles and Energy Flows  32  2.3.3 Principle Criteria no. 3 Presence of Functioning Recovery Mechanisms  35  -  -  -  2.3.5 Examination of Linkages Among Rangeland Health Indicators and Quantitative Vegetation and Soil Properties  36  2.4 Conclusions  37  2.5 References  39  3. Effects of Cattle Grazing on Aggregate Stability on Rough Fescue Grasslands in the Southern Interior of British Columbia  54  3.1 Introduction  54  3.2 Materials and Methods  55  3.2.1 Site Description  55  3.2.2 Sampling and Analyses  57  3.2.2.1 Soil Sampling  57  3.2.2.2 Plant Sampling  58  3.2.3 Statistical Analysis  59  3.3 Results and Discussion  60  3.3.1. Impacts of Grazing on Aggregate Stability  60  3.3.2 Relationships Among Soil and Vegetation Properties with Aggregate Stability.. 60 3.4. Conclusions  63  3.5 References  65  iv  4. enera1 onc1usions..  .  4.1 Research Conclusions  80  80  4.2 Evaluation of Study Methods and Recommendations for Future Research.. 82 4.3 References Appendices  86 88  V  List of Tables  Table 2.1  —  Location, site characteristics, and time of establishment of study sites in the southern  interior of British Columbia Table 2.2  —  43  Summary of regression analyses that correspond to the three principal criteria for  determination of rangeland health as outline by National Research Council (1994) (n=20) Table 2.3  —  44  Summary of indicators examined and their associations with the three principal  criteria for determination of rangeland health as outlined by the National Research Council (1994) at 20 rough fescue grassland treatment units in the southern interior of BC Table 2.4  —  45 Correlation among the top five indicators: Sandberg’s bluegrass, percent exposed  mineral soil, percent cover of Junegrass, percent litter cover, and percent cover of rough fescue Table 3.1  —  Location, site characteristics, and year of exciosure establishment of eight sites in the  southern interior of British Columbia Table 3.2  —  —  69  Soil textural classes and seral stage of 18 study treatment units located in the  southern interior of British Columbia Table 3.3  46  70  Soil aggregate stability parameters, total C and N, C:N ratio, soil polysaccharides,  soil bulk density, mechanical resistance, exposed mineral soil, and selected vegetation parameters in response to seral stage on 18 treatment units of rough fescue grasslands in the southern interior of British Columbia  71  vi  List of Figures Figure 2.01  —  Relationship between soil bulk density (Mg m ) and litter cover 3  (%) at 20 rough  fescue grassland treatment units in the southern interior BC Figure 2.02  —  47  Relationship between soil bulk density (Mg m ) and exposed mineral soil (%) at 3  20 rough fescue grassland treatment units in the southern interior BC Figure 2.03  —  47  Relationship between soil bulk density (Mg m ) and Koeleria macrantha cover 3  (%) at 20 rough fescue grassland treatment units in the southern interior BC Figure 2.04  —  Relationship between soil bulk density (Mg m ) and Poa secunda cover 3  47  (%) at 20  rough fescue grassland treatment units in the southern interior BC Figure 2.05  —  cover Figure 2.06  —  47  Relationship between soil mechanical resistance (kPa) at 4.5 cm depth and litter  (%) at 20 rough fescue grassland treatment units in southern interior of BC  48  Relationship between soil mechanical resistance (kPa) at 4.5 cm depth and exposed  mineral soil  (%) at 20 rough fescue grassland treatment units in southern interior of  BC Figure 2.07  48 —  Relationship between soil mechanical resistance (kPa) at 4.5 cm depth and  Sandberg’s bluegrass cover  (%) at 20 rough fescue grassland treatment units in southern  interior of BC Figure 2.08  —  48  Relationship between soil mechanical resistance (kPa) at 4.5 cm depth and rough  fescue cover  (%) at 20 rough fescue grassland treatment units in southern interior of  BC Figure 2.09  48 —  Relationship between fraction of total soil (kg kg ) in the 1-2 mm aggregate size 1  class and litter cover  (%) at 20 rough fescue grassland treatment units in the southern  interior of BC Figure 2.10  —  49  Relationship between fraction of total soil (kg kg’) in the 1-2 mm aggregate size  class and Sandberg’s bluegrass cover  (%) at 20 rough fescue grassland treatment units in  the southern interior of BC Figure 2.11  —  49  Relationship between fraction of total soil (kg kg’) in the 1-2 mm aggregate size  class and exposed mineral soil southern interior of BC  (%) at 20 rough fescue grassland treatment units in the 49  vii  Figure 2.12  —  Relationship between fraction of total soil (kg kg’) in the 2-6 mm aggregate size  class and timber milkvetch cover  (%) at 20 rough fescue grassland treatment units in the  southern interior of BC Figure 2.13  —  Relationship between total carbon  49  (%) at 0-7.5 cm depth and exposed mineral soil  (%) at 20 rough fescue grassland treatment units in the southern interior of BC 50 Figure 2.14 Relationship between total carbon (%) at 0-7.5 cm depth and Sandberg’s bluegrass cover (%) at 20 rough fescue grassland treatment units in the southern interior of BC....50 Figure 2.15 Relationship between total carbon (%) at 7.5-15 cm depth and litter cover (%) at —  —  20 rough fescue grassland treatment units in the southern interior of BC Figure 2.16  —  Relationship between total carbon  50  (%) at 7.5-15 cm depth and Junegrass cover (%)  at 20 rough fescue grassland treatment units in the southern interior of BC Figure 2.17  —  Relationship between total nitrogen  bluegrass cover  50  (%) at 0-7.5 cm depth and Sandberg’s  (%) at 20 rough fescue grassland treatment units in the southern interior  ofBC Figure 2.18  —  51 Relationship between total nitrogen (%) at 7.5-15 cm depth and litter cover  (%) at  20 rough fescue grassland treatment units in the southern interior of BC Figure 2.19  —  51  Relationship between total nitrogen (%) at 7.5-15 cm depth and exposed mineral  soil (%) at 20 rough fescue grassland treatment units in the southern interior of BC Figure 2.20  —  Relationship between total nitrogen n  51  (%) at 7.5-15 cm depth and Junegrass cover  (%) at 20 rough fescue grassland treatment units in the southern interior of BC 51 Figure 2.21 Relationship between soil polysaccharides (tg mr’) and Junegrass cover (%) at 20 —  rough fescue grassland treatment units in the southern interior of BC Figure 2.22  —  52  Relationship between soil polysaccharides (gig mf ) and common dandelion cover 1  (%) at 20 rough fescue grassland treatment units in the southern interior of BC Figure 2.23  —  52  Relationship between soil polysaccharides (gig m[’) and litter biomass (kg ha’) at  20 rough fescue grassland treatment units in the southern interior of BC Figure 2.24  —  Relationship between mean weight diamter (mm) and timber milkvetch cover  52  (%)  at 20 rough fescue grassland treatment units in the southern interior of BC Figure 2.25  —  Relationship between rough fescue seedheads (per m ) and rough fescue cover 2  at 20 rough fescue treatment units in southern interior of BC  52  (%) 53  viii  Figure 2.26  —  Relationship between bunchgrass density (greater than 4 cm) (per m ) and rough 2  fescue cover Figure 2.27  —  cover Figure 2.28  —  (%) at 20 rough fescue treatment units in southern interior of BC  Relationship between total aboveground living biomass (kg h&’) and rough fescue  (%) at 20 rough fescue treatment units in southern interior of BC Relationship between litter biomass (kg h&’) and rough fescue cover  53  (%) at 20  rough fescue treatment units in southern interior of BC Figure 3.01  —  Relationship between total carbon  1)  Relationship between total carbon  —  Relationship between total carbon  —  Relationship between total carbon  —  —  —  —  —  1)  —  —  74  Relationship between polysaccharides and the 2-6 mm aggregate size class (kg kg  at 18 rough fescue grassland treatment units in southern interior of BC  Figure 3.12  74  Relationship between polysaccharides and the 1-2 mm aggregate size class (kg kg  at 18 rough fescue grassland treatment units in southern interior of BC  Figure 3.11 1)  —  73  Relationship between polysaccharides and the 0.25-1 mm aggregate size class (kg  ) at 18 rough fescue grassland treatment units in southern interior of BC 1 kg Figure 3.10  73  Relationship between C :N ratio and mean weight diameter (mm) at 18 rough  fescue grassland treatment units in southern interior of BC Figure 3.09  73  Relationship between C:N ratio and the 2-6 mm aggregate size class (kg kg’) at 18  rough fescue grassland treatment units in southern interior of BC Figure 3.08  73  Relationship between C:N ratio and the 1-2 mm aggregate size class (kg kgj at 18  rough fescue grassland treatment units in southern interior of BC Figure 3.07  72  Relationship between C:N ratio and the 0.25-1 mm aggregate size class (kg kg’) at  18 rough fescue grassland treatment units in southern interior of BC Figure 3.06  72  (%) and mean weight diameter (mm) at 18  rough fescue grassland treatment units in southern interior of BC Figure 3.05  72  (%) and the 2-6 mm aggregate size class (kg kg  at 18 rough fescue grassland treatment units in southern interior of BC  Figure 3.04  72  (%) and the 1-2 mm aggregate size class (kg kg  at 18 rough fescue grassland treatment units in southern interior of BC  Figure 3.03 1)  —  53  (%) and the 0.25-1 mm aggregate size class (kg  kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.02  53  74  Relationship between polysaccharides and mean weight diameter (mm) at 18 rough  fescue grassland treatment units in southern interior of BC  74  ix  Figure 3.13  —  Relationship between bulk density (Mg m ) and the 0.25-1 mm aggregate size 3  class (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC...75 Figure 3.14  —  Relationship between bulk density (Mg m ) and the 1-2 mm aggregate size class 3  (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.15  —  75  Relationship between bulk density (Mg m ) and the 2-6 mm aggregate size class 3  (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.16  —  Relationship between bulk density (Mg m ) and mean weight diameter (mm) at 18 3  rough fescue grassland treatment units in southern interior of BC Figure 3.17  —  75  75  Relationship between mechanical resistance (kPa) and the 0.25-1 mm aggregate  size class (kg kg’) at .18 rough fescue grassland treatment units in southern interior of BC Figure 3.18  76 —  Relationship between mechanical resistance (kPa) and the 1-2 mm aggregate size  class (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC...76 Figure 3.19  —  Relationship between mechanical resistance (kPa) and the 2-6 mm aggregate size  class (kg kg ) at 18 rough fescue grassland treatment units in southern interior of BC...76 1 Figure 3.20  —  Relationship between mechanical resistance (kPa) and mean weight diameter (mm)  at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.21  —  Relationship between litter cover  (%) and the 0.25-1 mm aggregate size class (kg  kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.22 1)  Relationship between litter cover  77  (%) and the 1-2 mm aggregate size class (kg kg  at 18 rough fescue grassland treatment units in southern interior of BC  Figure 3.23 1)  —  76  —  Relationship between litter cover  (%) and the 2-6 mm aggregate size class (kg kg  at 18 rough fescue grassland treatment units in southern interior of BC  Figure 3.24  —  Relationship between litter cover  —  77  (%) and mean weight diameter (mm) at 18 rough  fescue grassland treatment units in southern interior of BC Figure 3.25  77  77  Relationship between Koeleria macrantha cover (%) and the 0.25-1 mm aggregate  size class (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.26  78 —  Relationship between Koeleria macrantha cover  (%) and the 1-2 mm aggregate  size class (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC  78  x  Figure 3.27  —  Relationship between Koeleria macrantha cover (%) and the 2-6 mm aggregate  size class (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.28  78 —  Relationship between Koeleria macrantha cover  (%) and mean weight diameter  (mm) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.29  —  Relationship between Poa secunda cover  78  (%) and the 0.25-1 mm aggregate size  class (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.30  79 —  Relationship between Poa secunda cover  (%) and the 1-2 mm aggregate size class  (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.31  79 —  Relationship between Poa secunda cover  (%) and the 2-6 mm aggregate size class  (kg kg’) at 18 rough fescue grassland treatment units in southern interior of BC Figure 3.32  79 —  Relationship between Poa secunda cover  (%) and mean weight diameter (mm) at  18 rough fescue grassland treatment units in southern interior of BC  79  xi  Acknowledgements I would first like to thank my supervisor, Maja Krzic, for her continuous support and guidance throughout the past two years. It has been a privilege to learn from and work for such an intelligent person. It has been a rough ride at times and without her encouragement and determination, I would not have done so well throughout my graduate studies. I would also like to give the sincerest of thanks to Reg Newman who has been a great teacher for me throughout my graduate studies. He was always there to give me any support or guidance that I may have needed. His intelligence and knowledge of rangeland plants and ecosystems has been overwhelmingly helpful and important for me to fully understand my research and expand my horizons past soil science. My committee members, Art Bomke and Gary Bradfield, have also been there to support me in every aspect of my research whether it was statistical analysis, giving feedback on my thesis, or just being able to sit in an office and talk about anything. I truly appreciate those days in Art Bomke’s office when I thought I couldn’t go any further. I do not think I would have had so much fun if it weren’t for my soil science student colleagues. My officemates, Melissa Iverson, Yihai (Simon) Zhao, and Stephanie Grand, kept me distracted from my thesis and shared many laughs and good times. I’d like to thank Jason Gray for his help in the field and in the laboratory. Also, I’d like to give a huge thanks to Les Lavkulich who provided laboratory and life assistance throughout this whole process. I will always remember the days spent in his office talking about random and not-so-random things. A very special thanks to Brian Wallace for his help in the field and for picking up many of my frantic phone calls. I would also like to thank the Forest Science Program and the University Graduate Fellowship for the funding and support of the project. Lastly, I would like to thank my parents for all their support (financial and emotional) throughout the many years of my studies. It is with their encouragement and guidance that I am able to be where I am today. I thank my mom for coming to visit me all those times when I needed someone and for my dad who knows exactly what it’s like to write a thesis. I’d also like to thank my brother and sister for all the phone conversations that helped me through so much of my life.  xii  Co-Authorship Statement Chapters 2 and 3 will be co-authored by Reg Newman, Brian Wallace, and Maja Krzic for publication.  Sarah Lamagna, thesis author, was responsible for the design of individual experiments, data collection, statistical analyses, and cooperation with the co-authors during the preparation of the manuscripts. Reg Newman provided guidance in experimental design, statistical analyses, data interpretation, and preparation of the manuscript. Brian Wallace was involved in sample collection and collaborated on the preparation of the manuscript in Chapter 2. Maja Krzic, research supervisor, initiated the research program, provided guidance through the project and during manuscript preparation.  xlii  1. General Introduction Grasslands are among the largest ecosystems in the world covering an estimated 52.5 million km 2 or 40.5% of the terrestrial area excluding Greenland and Antarctica (Suttie et al. 2005). Grasslands are ecosystems dominated by plants in the family Poaceae (grasses) with little or no woody plants (shrubs and trees) (Holechek et al. 1995). Typically, 50% of the plants in grasslands are grasses, while the other 50% is made up of forbs (Suttie et al. 2005). Another, more comprehensive definition of grasslands is offered by the Oxford Dictionary of Plant Sciences (Allaby 1998): “Grassland occurs where there is sufficient moisturefor grass growth, but where environmental conditions, both climatic and anthropogenic, prevent tree growth. Its occurrence, therefore, correlated with rainfall intensity between that ofdesert andforest and is extended by grazing and/orfire to form aplagioclimax in many areas that were previouslyforested.” Grassland distribution is affected by climate, disturbance (either by fire or grazing animals), and soil type (Suttie et al. 2005). Generally, coarse textured soils, low soil water content, limited precipitation, low-intensity fires, and wind are conditions that drive grassland distribution. Climate is the principal factor that promotes natural grasslands and the close similarity between the climatic and vegetational boundaries is usually quite apparent (Carder 1970). Grasslands generally occur in areas receiving between 250 and 900 mm of precipitation (Nebel 1981) by frequent light rains over an extended period (90 days or more). Carder (1970) described climatic factors that promote grassland development including continentality, drought, extended periods of cold, wind, and per humid conditions. Since grasslands are usually located far from the moderating effects of oceans, this causes a wide range of temperatures, with hot summers and cold winters and constant wind. The climate of the prairies in North America is influenced by their mid-continental location, and the sheltering effect of the Rocky Mountains. Grass can, on the whole, withstand this variability of climate better than trees. The reoccurrence of drought favours grassland conditions as well. Grasses are well-adapted to drought conditions and a moisture requirement is only critical in the spring and early summer when they are sprouting; however, there is some tendency for trees to encroach on grasslands (Carder 1970).  1  Fire is needed to overcome the natural tendency of some North American grasslands to be dominated by woody plants. No grassland with its associated plant community has developed without being influenced by fire (Vallentine 1989). Many prairie plants have adapted to fires by growing underground storage structures and having their growth points slightly below ground (Carder 1970). According to Vallentine (1989) there are three main reasons to burn grasslands: (i) to kill or suppress undesirable plants or shrubs, (ii) to prevent the occurrence of invasive species, and (iii) to increase forage production and thus grazing capacity. Depending on the geographical area and vegetation type, burning grasslands may promote the improvement of grazing use, release of plant nutrients from organic residue and litter, preparation of seedbeds for artificial seeding of forage species, or the temporary reduction of the amount of litter and vegetation that intercepts precipitation from light rains. Soil properties have a direct influence in grassland development. Grassland soils are usually deep, high in organic matter, and have high nutrient supply capacity. Surface horizons have an extensive fibrous root system of grasses, as well as numerous rhizomes, bulbs, and rootstock. The majority of grassland soils are either coarse or medium textured and consequently soil moisture is often limiting during parts of the growing season. Some grassland soils are saline, since their internal drainage is lower than the evapotranspiration. The most common grassland soils are Chernozems (or Mollisols according to the U.S. soil classification system) but other soil types may also occur on grasslands including Solonetz, Vertisol, Brunisol, and Regosol (Coupland and Brayshaw 1953; Suttie et al. 2005). Certain soils give rise to socalled edaphic grasslands in which soil conditions prevent tree growth in climate that would normally support forest ecosystems (Gayton 2003). Grasslands throughout the world (including those in the province of British Columbia) have been severely reduced by agricultural production and altered by inappropriate livestock grazing practices. Ongoing degradation of rangelands (including forests, wetlands, alpine communities, tundra, savannas, and other ecosystems grazed by domestic livestock) is a worldwide problem, currently affecting about 680 million hectares of rangelands (five times the United States cropland area). The extensive nature of this problem provides a strong rationale for studies that enhance our understanding of rangeland ecosystems and allow development of comprehensive monitoring strategies intended to improve management practices (USDA 2007).  2  1.1. North American Grasslands When European settlers first moved into the central portion of what is now North America, they encountered extensive grasslands extending from the prairies of Canada to the Gulf of Mexico (area now known as the Great Plains) (Pieper 2005). The Great Plains are relatively level, but minor topographic variations are important in influencing plant species distributions (Manning 1995). At coarse scales the Great Plains grassland of North America is commonly divided into tall-grass, mixed-grass, and short-grass. The grasslands in the central and northern portions of the Great Plains are a dynamic ecotone between the mountains and deserts to the west and the deciduous forests to the east (Pieper 2005). As one moves from east to west in North America, the rainfall in the prairies decreases. The environmental conditions are moister close to the mountains and to the east and north; they are driest in the central portions. Precipitation ranges from about 320 mm in the mixed prairie to 550 mm in the tall-grass prairie. Rainfall varies from year to year in the prairies; there is usually a long dry period during the summer months. Every 30 years or so there is a long drought period that lasts for several years (Suttie et al. 2005). The climate of the North American grasslands is relatively homogeneous in terms of its low winter rainfall, occasional droughts in summer and its tendency to occur synchronously within the region (Borchert 1950). In Canada, most grasslands occur within the four most western provinces. The natural grasslands of Canadian prairies can be divided into four main vegetation groups: tall-grass, mixed-grass, rough fescue, and palouse. Each prairie type has myriad of plant species and varies across the landscape depending on topography, moisture, soil type, fire frequency, grazing pressures, and competition. Tall-grass prairie can be found in south central Manitoba, southwestern Ontario, and south to Texas in the United States while mixed-grass prairie can be found in southern Manitoba and branching out westwards into Alberta. Rough fescue prairie is found in Riding Mountain National Park, Manitoba; Prince Albert National Park, Saskatchewan; and Waterton National Park, Alberta; as well as Cypress Hills, Alberta/Saskatchewan; and the interior of British Columbia. Lastly, the palouse prairie is found in British Columbia and extends as far south as Utah into the United States (Malin 1967). Within the province of British Columbia there are two types of grasslands  —  rough fescue  and palouse prairie (Campbell and Bawtree 1998). British Columbia’s grasslands are among the most diverse in North America because of the wide variations in climate, soils, topography, and  3  historical use (Tisdale 1947). This exceptional diversity makes classification and management of the grasslands very challenging. To provide a conceptual framework for management of forests and grasslands throughout British Columbia, the Ministry of Forests in 1975 initiated the development of the biogeoclimatic ecosystem classification (BEC). There are 14 BEC zones that are based on the dominant climax species present along with other factors such as soil moisture, slope, soil depth and texture, and topographic location. Eleven of the 14 BEC zones are used for domestic livestock grazing (Wikeem et al. 1993). Out of the 94 million hectares of land area in British Columbia, the ranching industry uses 10 million hectares (8.5 million hectares is Crown land). Grasslands make up about 10% of British Columbia’s rangelands (Wikeem eta!. 1993), while the remaining 90% is under either open forest or occur in clearcuts (with and without forage seeding). British Columbia’s grasslands are unique ecosystems, consisting of species and habitats that contrast sharply with the North American prairies. Grasslands of British Columbia contain more than 30% of the provincial wildlife species of concern, and support more threatened or endangered species than any other ecosystem. This is in agreement with a general recognition that grasslands are critical for maintenance of global biodiversity (Gayton 2003). Yet, between 76 and 99% of British Columbia’s native grasslands have been altered either due to agricultural production, inappropriate livestock grazing practices, urban expansion, invasive weeds, abusive recreational activities, forest encroachment, or climate change (Campbell and Bawtree 1998). Consequently, sustainability of grasslands is important for both biodiversity maintenance and the ranching industry. British Columbia’s grassland regions have also been recognized internationally for recreation and tourism opportunities. 1.1.1. Rough Fescue Prairie  Rough fescue prairie occupies a moister environment than does the mixed-grass prairie and is characterized by greater species abundance, richness, and productivity than the mixed grass prairie. On average, rough fescue prairie will produce twice as much forage as the most productive mixed-grass prairie, but rough fescue can easily be eliminated if overgrazed. Rough fescue prairie is not as threatened by grazing as tall-grass prairie but overgrazing is still of concern. Because of their productive Black Chernozemic soils, rough fescue grasslands were among the first areas to be converted to farmland during settlement of the west. Once extending  4  over 255,000 km 2 in the Prairie Provinces of Canada, less than 5% of the original rough fescue prairie remains (Coupland and Brayshaw 1953). The rough fescue prairie grasslands are largely restricted to Canada along the transition zone between the northern forests and the Great Plains. There are three distinct types of rough fescue: (i)’Northern’ rough fescue (Festuca altaica Trill.) mostly found in grasslands of northern British Columbia, Yukon, and Alaska; (ii)’Plains’ rough fescue (Festuca hallii (Vasey) Piper) found in the western plains and parkland in western Manitoba, Saskatchewan, Alberta, and southwestern Ontario; and (iii) ‘Mountain’ rough fescue (Festuca campesfris Rydb.) that occurs in the grasslands of southern British Columbia and Alberta (Brink 1982). 1.1.2. Palouse Prairie Palouse prairie grasslands are found only in the southern interior of British Columbia in Canada as well as in the Northwest part of the United States extending as far south as Utah. These grasslands can be split into two different types: the Festuca and the Agropyron. The Agropyron is dominated by bluebunch wheatgrass (Pseudoroegneria spicata (Pursh) A. Love syn. Agropyron spicatum), while the Festuca is dominated by Idaho fescue (Festuca idahoensis Elmer) and occurs in moister areas (Stoddart 1941). Characteristic plant species of the palouse prairie include mountain rough fescue, bluebunch wheatgrass, Columbia needle grass (Achnatherum nelsonii (Scribn.) Barkw.), and Kentucky bluegrass (Poapratensis L.).  1.2. Impacts of Grazing on Plant Communities Grazing can impact plant communities directly by defoliation and trampling of plants, or indirectly through changes in the energy balance at the soil surface, creation of different levels of systems disturbance, alteration of plant colonization, and redistribution of nutrients (Trlica and Rittenhouse 1993). The individual direct and indirect responses of plants to grazing animals ultimately controls the composition of the plant community (Holechek et al. 1995). Plant species have different responses to grazing because there are considerable differences in resistance to grazing among plant species (Caldwell et al. 1981). Grazing resistance can be divided into avoidance and tolerance components, based on the general mechanisms conferring resistance. Grazing avoidance involves mechanisms that reduce the probability and severity of grazing, while grazing tolerance consists of mechanisms that promote growth following defoliation. Avoidance mechanisms are composed of architectural 5  attributes, mechanical deterrents and biochemical compounds, which reduce tissue accessibility and palatability. Tolerance mechanisms are composed of the availability and source of residual meristems and physiological processes capable of promoting growth (Briske 1996). There are several factors that increase grazing resistance in grasses, forbs, and shrubs. Grasses have a higher proportion of culmiess shoots than other life forms, have a greater delay in elevation of the apical buds, sprout more freely from basal buds after defoliation, and have a higher ratio of vegetative to reproductive stems. Forbs produce a large number of viable seeds, have a delayed elevation of growing points, and have poisons and chemical compounds that reduce palatability. Shrubs have spines and thorns that discourage browsing, volatile oils and tannins that reduce palatability, have branches that make removal of inner leaves difficult, and only the current year’s growth of most shrub species is palatable and nutritious (Holechek et al. 1995). Grazing resistance in plants can occur through mechanical and biochemical mechanisms (alkaloids, glucosinolates, and cyanogenic substances) as well as through physiological and morphological mechanisms (Briske 1991). Mechanical and chemical mechanisms operate through plant accessibility and palatability to specific herbivores. Tissue accessibility is primarily a function of degree of elevation of leaves and tillers above the soil surface. Species without culmed shoots such as blue grama (Bouteloua gracilis (Willd. ex Kunth) Lag. ex Griffiths) are particularly resistant to defoliation because apical meristems are near the soil surface where accessibility is low. Spines, awns, and other epidermal characteristics (pubescence, silica, and cuticular wax) make plants unpleasant to touch and directly reduce palatability (Briske 1996). 1.2.1 Succession of British Columbia Grasslands Traditionally, succession was considered a progressive, unidirectional change in the composition of a vegetation community on a particular site. The quantitative climax approach to rangeland management and assessment has focused attention on successional processes (Lauenroth and Laycock 1989). Excessive grazing is perceived to lead to retrogression or decline in rangeland condition (Stoddart et at. 1975; Rodriguez and Kothmann 1997) and it is assumed that reduction or removal of grazing pressure allows successional processes to restore the range to what it was, essentially by reversing the path taken by retrogression  6  (Friedel 1991). The beginning of succession (or last stage of retrogression) was bare soil while the end point was a single “climax” set of native species. Along the way to climax, different sets of species are present among the ecosystem, which is known as seral stages. The beginning of a vegetation community starts with early-seral, pioneer species and the climax of a vegetation community consists of late-seral native species (Gayton 2003). There are three classes of plants based on their response to increasing levels of disturbance (i.e., grazing). The classes are (i) decreasers, (ii) increasers, and (iii) invaders. Decreasers are those palatable, late seral grasses and forbs that decrease in dominance or even disappear as grazing pressure increases. Early seral increasers are generally unpalatable and tend to increase as grazing pressure increases. Invaders are introduced or weedy species that appear after grazing pressure has weakened the existing native plant community (Dyksterhuis 1949). Three general seral stages can be seen in the rough fescue grasslands of the southern interior of British Columbia. The early seral stage of rough fescue grasslands consist of invasive species (i.e., diffuse knapweed and spotted knapweed) but very little or no rough fescue. The mid seral stage of rough fescue grasslands consist of a few types of bunchgrasses (i.e., Idaho fescue, bluebunch wheatgrass) and low cover of rough fescue. The late seral stage of rough fescue grasslands consist of higher bunchgrass cover especially rough fescue but also including bluebunch wheatgrass and Idaho fescue (Gayton 2003).  1.3. Impacts of Grazing on Soil Properties Grazing animals impact soil physical, chemical, and biological properties by trampling, redistribution of feces, and consumption of plants. The nature and degree of grazing impact on soil properties will depend on the type of grazing system, the stocking rate, and the season of grazing. Changes in the soil are important indicators of site deterioration. Rangeland monitoring must incorporate both the herbaceous layer as well as the soil (Friedel 1991). 1.3.1. Soil Physical Properties Soil physical properties that are most directly impacted by grazing include bulk density, strength, aggregate stability, pore size distribution, water infiltration rate, and water holding capacity.  7  1.3.1.1. Soil Compaction Soil compaction is defined as the process of packing together of soil particles by forces exerted at the soil surface increasing the soil bulk density. The most common soil properties that determine susceptibility to compaction are texture, structure, quantity and quality of organic matter, and water content. Stationary pressures on many soils do not exceed the supporting capacity of soils; however, as livestock walk their weights fall on restricted areas on their hooves, whereby weight per unit area does exceed the soil strength making it very susceptible to compaction (Heady and Child 1994). The result can be chipping of dry soil, compaction of moist soil, and deformation of wet soil (Ganjegunte et al. 2005). Generally, higher stocking rate increases soil bulk density, decreases macroporosity, and destroys aggregates (Vallentine 2001). Abdel-Magid et at. (1987) studied the impact of three grazing systems (continuous, rotationally deferred, and short-duration rotation) and stocking rate (heavy-2.25 ha steer’ and moderate-3.0 ha steer ) on soil bulk density and infiltration rate of a 1 sandy loam soil at the High Plains Grasslands Research Station in Cheyenne, Wyoming. Soil bulk density was not significantly affected by grazing system or stocking rate. The bulk density was significantly greater in the fall than in the spring, most likely as a result of freezing and thawing effect in the soil during the spring time (Abdel-Magid et at. 1987). In another study carried out on a coarse-textured soil in Brandon, Manitoba, Banerjee et at. (2000) showed that soil bulk density was significantly affected by the grazing (stocking) rates and as stocking rate increased (from 1.1 steers ha’ to 2.2 steers ha’) the soil bulk density also increased. When proper management was practiced and levels of stocking rates were maintained at an appropriate level, there were minimal grazing impacts on soil bulk density. 1.3.1.2. Soil Structure Soil structure has profound influence on overall ecosystem productivity (Mapfumo et al. 2000) since it has a direct impact on pore size distribution, which in turn impacts soil water dynamics (by partitioning the rainfall at the soil surface into runoff and infiltration), root growth, the exchange of gasses in the root zone, the physical habitat for soil biota, and the energy required for root penetration and ground engaging tools (Naeth et at. 1991). Macropores present among large aggregates enhance water infiltration rate (and reduce runoff). On the other hand, presence of unstable aggregates (easily disrupted by raindrop impact and trampling by cattle) at the soil surface may lead to clogging of macropores leading to reduction of infiltration. 8  Soil aggregates can be destroyed resulting from compaction and vibration of hooves (Vallentine 2001). Livestock trampling under intensive rotation grazing (2.7 animal unit months -  AUM) on a silty clay soil was tested at the Texas Agricultural Experiment Station near Sonora,  Texas to determine the effects on soil hydrology (Warren et al. 1986). Part of the study focused on aggregate size distribution and stability and their relation to soil hydrologic response of the trampling treatments. Four rates of trampling were applied to a small paddock in the study area and included moderate (8.1 ha/AU/yr), double (4.1 ha/AU/yr), triple (2.7 haJAU/yr), and no trampling. The mean infiltration rate across all trampling intensities was significantly higher from plots, which were trampled dry than from plots which were trampled moist (Warren et al. 1986). Warren et al. (1986) concluded that size and stability of soil aggregates were generally closely related to infiltration rate. Interspaces between large aggregates increase the macroporosity of the soil which, in turn, enhances infiltration rate. As infiltration rate increases, less runoff is available to transport sediment. In addition, large aggregates are less likely to be transported by surface runoff water. Aggregate size distribution was negatively correlated to infiltration rate and positively correlated to sediment production on soil, which was trampled while moist. This was caused by the formation of large comparatively impermeable clods when the soil was compacted while moist (Warren et al. 1986). 1.3.1.3. Water Infiltration Rate and Soil Water Content As soil bulk density increases due to compaction by cattle, infiltration and percolation decrease. Abdel-Magid et al. (1987) evaluated the infiltration rates of soils under three grazing systems (continuous, rotationally deferred, and short-duration rotation) and three stocking rates (2.25 ha steer, 3.0 ha steer, and 5.25 ha steerj. Both equilibrium and cumulative infiltration were affected by a year-by-grazing system and a season-by-stocking rate interaction. The continuous grazing treatment resulted in a significantly higher average equilibrium infiltration rate (9.69 cm hrj than the rotational deferment treatment (8.13 cm hr ). The average water 1 infiltration was significantly lower in the fall than in the spring for the heavy stocking rate. No significant differences in infiltration among stocking rates existed in the spring, indicating that the freeze-thaw activity each winter alleviated any detrimental soil compaction that reduced infiltration. The heavy stocking rate reduced infiltration in one of the two years tested compared to the moderate level of stocking. Stocking rate seems consistently to be a more important  9  influence on infiltration rate and bulk density than is the type of grazing system (Abdel-Magid et at 1987). Soil, as well as the vegetation growing thereon, has substantial resiliency that permits it to overcome many short-term effects of trampling. Soil compaction from grazing often disappears or decreases after seasonal wetting and drying or freezing and thawing. Although related to infiltration rates, increased compaction does not necessarily result in lower soil water content because the effects of grazing on reducing evapotranspiration may be even greater (Vallentine 2001). In a study on the grazing effects on three different ecosystems in southern Alberta, Naeth et al. (1991) found the amount of large-sized particles relative to small-sized particles litter and organic matter were main factors in determining water holding capacity in a given rangeland. They found that water holding capacity increased with increasing litter particle-size. Water holding capacity decreased with heavy intensity, early season grazing through change in species composition and cattle trampling. Thus, grazing regimes facilitating accumulation of litter and larger particle-sized categories would increase water holding capacity (Naeth et al. 1991). In the fescue grasslands of the Alberta foothills, there were obvious differences in soil water content between the ungrazed control site and the grazed (heavy = 2.4 AUM ha 1 and very heavy = 4.8 AUM ha) sites (Naeth and Chanasyk 1995). Surface soil water (0-7.5 cm) across 1 slope positions was lowest in the control and highest in the very heavily grazed area, but the trend in the profile soil water (to 50 cm) was the opposite. The “undisturbed” soil with greater vegetation cover allowed less surface runoff and consequently there was a greater water infiltration and percolation relative to both grazed areas. Both grazed areas had less vegetation and allowed greater surface runoff. The vegetation at the study site was regularly water-stressed (more so in both grazed areas); soil water was often below permanent wilting point by mid summer. Soil water was generally near or above field capacity every spring, indicating the importance of snowmelt infiltration in these ecosystems (Naeth and Chanasyk 1995). 1.3.2. Soil Chemical Properties The length of time that grazing animals spend in certain parts of rangelands may also determine type and intensity of impact on soil properties. Both manure and urine excreta of freely grazing animals are deposited most heavily where animals spend the most time rather than  10  where the forage is produced and consumed. Thus, forage producing parts of the rangeland become progressively more deficient in plant nutrients while areas near water, salt, feeding areas (conserved feed), bed grounds, and shade are enriched with plant nutrients (Vallentine 2001). 1.3.2.1. Soil Organic Matter Preferential grazing, trailing, and camping behaviour of livestock act to redistribute nutrients and organic matter contained in both dung and urine (Packer 1988). Grazing by large herbivores increases nutrient cycling rates by reducing the size of organic particles (increasing total surface area and the amount of reactions that can take place) but also by accelerating the rate of nutrient conversion from an organic to inorganic form available to plants. Fecal N is largely insoluble and becomes available to plants only after incorporation into the soil and decomposition carried out by soil fauna and microorganisms. The N in urine is readily available or rapidly becomes so, the proteins and amino acids are converted to nitrate and ammonium (Vallentine 2001). According to Woodmansee (1978), about 17% of ingested forage N remained in tissues of domestic ungulates (the amount varied with carrying capacity and degree of utilization of forage), while the remainder of ingested forage was excreted and became the principal pooi from which N was lost from grasslands. Losses via volatilization of NH 3 from ungulate urine can range as high as 80 to 90% of the N initially therein, and losses from feces may be 20% or higher. Assuming 35% of the excreta deposited by ungulates fell within 10% of the pasture (near stock tanks, fences, and bedding areas), N was essentially removed from circulation within the pastures. Many grazing animals either excrete NH 3 directly, or they excrete nitrogenous compounds that readily hydrolize to NH . Presumably, all NH 3 3 generated below several centimetres of the soil surface remains in the system, while ammonia produced at or near the soil surface is more vulnerable to loss. Since many soil organisms are active near the soil surface and in the litter at the surface, losses from these sources may be significant for grassland ecosystems (Woodmansee 1978). Poor grazing management can result in contamination of surface and subsurface waters directly through bacterial contamination, nutrient over-enrichment, and indirectly by soil erosion from pastures. Recent research on the effects of fertilizer, manure, and urine on N cycling provides some interesting information. For example, a study by Ganjegunte et al. (2005) carried  11  out for one year in pastures receiving 200 to 250 kg N ha , found that N that was returned to the 1 soil surface as urine or manure ranged from 154 kg N ha’ for beef steers to 300 kg N ha’ for dairy cows. In grass-clover pastures receiving no N fertilizer, the value was slightly less at 132 kg N ha’. Considering that a manure pile covers around 0.01 m 2 and a urine spot covers 0.3 70.65 m , it is interesting to note that the soil under each dairy cow manure pile or urine spot 2 receives the equivalent of 560 to 1120 kg N ha’ (Ganjegunte et al. 2005). This can cause severe changes in the soil, directly affecting the species composition of the land, which will in the long term affect the livestock itself (Henderson et al. 2004). The N in urine is immediately available to the plants, and most plants are overwhelmed with the high N concentration around a urine spot. Some of this N may be lost to the atmosphere. In contrast, the N in manure is released more slowly than the N in urine. The ability of plants to take up this much N is limited; they simply cannot use such heavy rates of N efficiently (Mapfumo et al. 2000). The quantity and quality of soil organic matter is an important factor in determining soil quality (Gregorich et al. 1994). Soil organic matter contributes to increase both cation exchange capacity and water-holding capacity. It also contains large quantities of some plant nutrients (e.g., N, 5, and P). Certain components of soil organic matter (e.g., polysaccharides) are largely responsible for the stabilization of soil aggregates. The effects of grazing on litter and soil organic matter were evaluated by Naeth et al. (1991). The study was done on three different ecosystems (mixed prairie, parkiand fescue, and foothills fescue) in southern Alberta. Several grazing treatments were studied during different seasons. In mixed prairie, a heavy stocking rate (4.4 AUM ha ) was applied and three grazing 1 treatments were studied: early season grazing, late season grazing, and ungrazed. In parkland fescue, five grazing treatments were applied: June grazing at 1.5 AUM ha’, heavy June grazing at 4.4 AUM ha , heavy autumn grazing at 4.4 AUM ha’, light autumn grazing at 1.5 AUM ha’, 1 and a control ungrazed. The foothills fescue had five grazing treatment also: very heavy grazing at 4.8 AUM ha , heavy grazing at 2.4 AUM ha’, moderate grazing at 1.6 AUM ha’, light 1 grazing at 1.2 AUM ha , and a control comprised of permanent exclosures in each treatment. 1 Both season and intensity of grazing affected amounts of litter and organic matter. Bare ground increased while standing and fallen litter, and live vegetative cover and mass decreased with increasing grazing intensity in all three ecosystems. Early season grazing was more detrimental than late season grazing. Litter depth and plant height were reduced with treading through  12  breakage and compaction. Proportions of different particle-sized litter varied with grazing season and intensity, although biomass was affected more consistently by intensity rather than season of grazing (Naeth et al. 1991). Early in the growing season when plant growth is rapid and carbohydrate reserves are low, vegetation is more susceptible to grazing damage. Heavy intensity grazing removes more vegetation with less regrowth and litter accumulation. Grazing also can affect litter decomposition rate, which in turn affects soil organic matter content. Trampling can reduce litter particle size and create better litter-soil contact, facilitating more rapid decomposition by soil microorganisms in some grazed treatments than in controls. This was observed in a similar study by Naeth et al. (1991) which showed lower microbial biomass and total organic matter content in the controls than in the light autunm treatment in parkland fescue and the late season treatment in mixed prairie. They also found higher mass of very fine size litter particles in very heavy and heavy grazing treatments than in light and moderate treatments in the foothill fescue. This shows that grazing enhances decomposition rates (Naeth et at. 1991). The increases of bare ground area in mixed prairie and parkiand fescue under grazing observed in the study by Naeth et at. (1991) were of little practical significance because the increases were so small (about 2%). Increases in bare ground (almost 25%) in foothills fescue under heavy and very heavy grazing rates were of practical significance since hydrologic changes (i.e., reduced infiltration and increased runoff) occurred in this ecosystem when bare ground was about 15% (Naeth et al. 1991). 1.3.2.2. Soil Polysaccharides Carbohydrates comprise about 10-15% of the total soil organic carbon. Most of the carbohydrates are present as a mixture of various monosaccharides (or simple sugars), oligosaccharides, and polysaccharides. Soil polysaccharides are comprised of a wide range of monosaccharides with simple and complex molecular structure and are formed in soil either as a result of plant root activity or microbial activity during decomposition processes (Haynes and Swift 1990; Tisdale and Oades 1982). There are a number of effects soil polysaccharides have on soil properties but they are most importantly related to the formation and stabilization of soil structure, and providing a readily available source of energy for soil organisms.  13  Soil carbohydrates in general, and polysaccharides in particular, have been studied in relation to soil aggregation. Some studies have found good correlation between carbohydrates content and soil macroaggregate stability (Haynes and Swift 1990; Angers and Mehuys 1993; Martens and Frankenburger 1992), while others have not (Carter et aL 1993). The most likely reason for lack of agreement among the studies is that other soil organic matter compounds such as the hydrophobic aliphatic fraction (Capriel et al. 1990) and fungal hyphae (Tisdale and Oades 1982) are involved in macroaggregate stability. Another reason for opposing findings reported in the literature is related to the lack of standard extraction and hydrolysis procedures used to determine soil carbohydrate and polysaccharides (Gregorich et al. 1994). A study done in the foothills of southwestern Alberta by Dormaar and Willms (1998) evaluated the effects of grazing on monosaccharide content. Paddocks were stocked at rate of 1.2 (light), 2.4 (heavy), and 4.8 (very heavy) animal unit month or AUM ha’ since 1949. They found that overgrazing reduced the amount of monosaccharides and a potential energy source in the soil which reduced microbial activity in the soil.  1.4 Rangeland Health Assessment Methods The plant community approach to assessing rangeland condition came into question in the 1990s (SRM Task Group 1995) and rangeland health assessments were proposed as replacements for the traditional system based on plant seral stage of the site. Rangeland health assessment methods use an holistic approach to determine the state of rangelands, and include assessments of ecological functions in addition to plant community information. Rangeland health has been defined by Adams et al. (2005) as “The ability ofrangeland to perform certain key functions ...where all parts that make up the whole are present and working together,” while Busby (1994) gave another definition “Rangeland health is the degree to which the integrity of the soil and the ecological processes ofrangeland ecosystems are sustained.” Most rangeland health assessment methods have been designed to be simpler than traditional methods of rangeland condition assessment, allowing wider use with minimal training. Visual rating of indicators are often used in favour of quantitative measures. Because of the reliance on visual ratings, there are some concerns regarding the reliability of rangeland health assessments. In British Columbia, two commonly used rangeland health assessment methods are the BC  14  Ministry of Forest and Range upland assessment method (British Columbia Ministry of Forests and Range 1997) and the Grassland Conservation Council method for the fescue grasslands (Wikeem and Wikeem 2005). To our knowledge, no studies have evaluated the linkages among indicators of rangeland health assessments and quantitative measures of plant and soil properties in British Columbia and only a limited number of studies have focused on quantification of these relationships anywhere in the world. A recently published manuscript by Miller (2008) evaluated the status of three ecosystem attributes (soil/site stability, hydrologic function, and biotic integrity) using the qualitative assessment protocol “Interpreting Indicators of Rangeland Health” across a broad spatial extent of 760,000 ha in Grand Staircase  —  Escalante National Monument, Utah. This  study found that sites with the greatest production potential were the most degraded and that net effects of past management practices have not been ecologically beneficial, indicating that ongoing management needs to be site specific accounting for site sensitivity to degradation. Since rough fescue grasslands on British Columbia are very sensitive to grazing, they are well suited for studies that will focus on the establishment of relationships between rangeland health assessment indicators and soil and vegetation properties. The National Research Council (1994) in the United States discussed recommendations for new methods of assessing rangeland as did the Society for Range Management Task Group on Unity in Concepts and Terminology Committee (1995). Both groups pointed out that rangeland assessments should focus on indicators of soil stability, watershed function, nutrient cycling, energy flow, and recovery mechanisms; although all these indicators of rangeland health are important, the focus must primarily be on soil conservation. Assessments should evaluate rangeland plant communities in terms of their ability to protect a site against accelerated soil erosion (Miller 2008).  1.5 Summary of General Introduction Distribution of grasslands is affected by climate, disturbance (either by fire or grazing animals), and soil type. Grasslands are generally present on coarse textured, dry soil, in regions with a limited precipitation and low-frequency fires. In Canada, most grasslands occur within the four most western provinces and are divided into four main vegetation groups: tall-grass,  15  mixed-grass, rough fescue, and palouse. In British Columbia, rough fescue and palouse prairie are the dominant grassland types. There are numerous impacts (both positive and negative) of livestock grazing on soil properties and plant communities on grasslands. Various plant species have adapted to the stress of grazing by creating defence mechanisms or having a high stress tolerance. Defence mechanisms include methods that reduce the probability and severity of grazing (e.g., thorns, spines, chemicals), while grazing tolerance consists of mechanisms that promote growth following defoliation. The negative effects of grazing include an increase in soil bulk density and subsequent decrease in infiltration and percolation. Grazing, depending on the stocking rates and the season of use, can be a healthy alternative compared to more intensive land uses (e.g., intensive agriculture, recreation use). To assess grazing management, various health assessment methods are used to create visual rating of indicators. An holistic approach has replaced the traditional system of using plant seral stage as a health assessment. Despite the widespread adoption and increasing use world wide, these new rangeland health assessment systems need to be evaluated by quantifying linkages among indicators of rangeland health assessment methods and quantitative measures of plant and soil properties.  16  1.6 Study Objectives Sustainable grassland ecosystems can only be maintained under optimal grazing management. The overall goal for this study was to evaluate soil quality and plant composition of rough fescue grasslands with and without long-term grazing within the Interior Douglas-fir biogeoclimatic zone in the southern interior of British Columbia. This overall goal was addressed through the following two study objectives:  Objective 1  —  to quantify the relationships among soil/vegetation properties known to be affected  by grazing to easily-assessed indicators, used in the existing health assessment systems, that do not require laboratory analyses or time consuming measurement. Objective 2 to evaluate impacts of grazing on soil aggregate stability on the rough fescue -  grasslands of the southern interior of British Columbia.  The hypothesis tested under objective 1 was: 1) Soil and plant properties, known to be affected by grazing, are related to easily-assessed indicator variables used in health assessment systems. The hypotheses tested under objective 2 were: 1) Grazing reduces soil aggregate stability, total C and N, and polysaccharides, while increasing bulk density and mechanical resistance, relative to the rough fescue grasslands without grazing. 2) Aggregate stability is positively related to total C, C:N ratio, total soil polysaccharides, litter biomass, and litter cover. 3) Aggregate stability is negatively related to soil bulk density, soil mechanical resistance, and exposed mineral soil. The long-term objective of the study is to advance the conceptual understanding of the grazing effects on soil and vegetation properties, which would in turn allow for refinement of existing indicators and assessment systems of grassland health.  17  1.7 References Abdel-Magid, A.H. G.E. Schuman, and R.H. Hart. 1987. Soil bulk density and water infiltration as affected by grazing systems. Journal of Range Management 40: 307-309. Adams, B.W., G. Ehlert, C. Stone, D. Lawrence, M.Alexander, M. Willoughby, C.Hincz, D. Moisey, and A. Bogen. 2005 Rangeland Health Assessment for Grassland, Forest and Tame Pasture. Alberta Sustainable Resource Development. Public Lands Division. Edmonton, AB. Agresti, A. 1996. An introduction to categorical data analysis. John Wiley & Sons, New York. Allaby, M. 1998. Oxford Dictionary of Plant Sciences. Oxford University Press, Oxford, UK. Andersen, A.N. 1990. The use of ant communities to evaluate change in United States terrestrial ecosystems: a review and recipe. Proceedings of the Ecological Society of United States 16: 347-357. Angers, D.A. and G.R. Mehuys. 1993. Aggregate stability to water. p. 65 1-658. In M.R. Carter (ed.) Soil sampling and methods of analysis. Lewis Publisher, Boca Raton, FL. Banerjee, M.B., D.L. Burton, W.P. McCaughey, and C.A. Grant. 2000. Influence of pasture management on soil biological quality. Journal of Range Management 53: 127-132. Borchert, J.R. 1950. The climate of the Central North American grassland. Annals Assoc. Amer. Geog. 40:1-39. Brink, V.C. 1982. An overview of the grasslands of North America. p. 27-38. In Nicholson, A.C., A. McLean, and T.E. Baker (eds). Grassland Ecol. Class. Symp. Proc., Kamloops, B.C. Briske, D. 1991. Developmental morphology and physiology of grasses. p. 27-33. In R.K. Heitschmidt and J.W. Stuth (eds). Grazing Management: An Ecological Perspective. Timber Press, Portland, Ore. Briske, D. 1996. Strategies of plant survival in grazed systems: a functional interpretation. p. 3 7-67. In Hodgson, J. and A.W. Illius (eds). The Ecology and Management of Grazing Systems. CAB International. Texas. British Columbia Ministry of Forests and Range. 1997. Range and riparian assessment reference manual. Ministry of Forests, Forest Practices Branch, Range section, Victoria, B.C. Caidwell, M.M, J.H. Richards, D.A. Johnson, R.S. Nowak, and R.S. Dzurec. 1981. Coping with herbivory: Photosynthetic capacity and resource allocation in two semiarid Agropyron bunchgrasses. Oecologia, 50: 14-24.  18  Campbell, C.W. and A.H. Bawtree. 1998. Rangeland Handbook for BC. British Columbia Cattleman’s Association. Kamloops, BC. Capriel, P., T. Back, H. Borchert, and P. Harter. 1990. Relationships between soil aliphatic fraction extracted with supercritical hexane, soil microbial biomass and soil aggregate stability. Soil Science Society of America Journal 54: 4 15-420. Carder, A.C. 1970. Climate and the rangelands in Canada. Journal of Range Management 23: 263-267. Coupland, R.T. and T.C. Brayshaw. 1953. The fescue grassland of Saskatchewan. Ecology 34: 386-405.  Dyksterhuis, E.J. 1949. Condition and management of rangeland based on quantitative ecology. Journal of Range Management 2: 104-115. Friedel, M.H. 1991. Range condition assessment and the concept of thresholds: a viewpoint. Journal of Range Management 44: 422-426.  Ganjegunte, G.K., G.F. Vance, C.M. Preston, and G.E. Schuma. 2005. Soil organic carbon composition in a northern mixed-grass prairie: effects of grazing. Soil Science society of America Journal 69: 1746-1756. Gayton, D.V. 2003. British Columbia grasslands monitoring vegetation change. F. Research Extenstion Partnership, Kamloops, B.C. Forrex Series 7. —  Gregorich, E.G., M. R. Carter, D.A. Angers, C.M. Monreal, B.H. Ellert. 1994. Towards a minimum data set to assess soil organic matter quality in agricultural soils. Can J. Soil Sci. 74: 367-385. Haynes, R.J. and R.S. Swift. 1990. Stability of soil aggregates in relation to organic constituents and soil water content. Eur. J. Soil Science. 41: 73-83. Heady, H.F. and D.R. Child. 1994. Rangeland ecology and management. Westview Press. Boulder, CO. Henderson, D.C., B.H. Ellert, and M.A. Naeth. 2004. Grazing and soil carbon along a gradient of Alberta rangelands. Journal of Range Management 57: 402-4 10. Holechek, J.L., R.D. Pieper, and C.H. Herbel. 1995. Range Management: principles and practices. Prentice Hall, New Jersey. Lauenroth, W.K. and W.A. Laycock. 1989. Secondary succession and the evaluation of rangeland condition. Westview Press, Boulder, CO.  19  Maim, J. 1967. The grassland of North America. P.Smith, Glaucester, MA.  Mapfiimo, E.M., D.S. Chanasyk, V.S. Baron, and M.A. Naeth. 2000. Grazing impacts on selected soil parameters under short-term forage sequences. Journal of Range Management 53: 466-470. Martens, D.A. and W.T. Frankenberg Jr. 1992. Decomposition of bacterial polymers in soil and their influence on soil structure. Biology of Fertile Soils 13: 65-73. Miller, M.E. 2008. Broadscale assessment of rangeland health, Grand Staircase-Escalante National Monument, U.S.A. Rangeland Ecology and Management 6 1:249-262. Naeth, M.A and D.S. Chanasyk. 1995. Grazing effects on soil water in Alberta foothills fescue grasslands. Journal of Range Management 44: 528-534. Naeth, M.A., A.W. Bailey, D.J. Pluth, D.S. Chanasyk, and R.T. Hardin. 1991. Grazing impacts on litter and soil organic matter in mixed prairie and fescue grassland ecosystems of Alberta. Journal of Range Management 44: 7-12. National Research Council. 1994. Rangeland health. New methods to classify, inventory and monitor rangelands. NRC., Washington, D.C. Nebel, B.J. 1981 Envrionmental Science: the way the world works. Prentice Hall Inc. Englewood Cliffs, NJ. Pieper, R.D. 2005. Grasslands of central North America p. 22 1-263 In S.G. Reynold and C. Batello (eds) Grasslands of the World in the Plant Production and Protection Series. Rome, Italy Rodriguez, R.M. and M.M. Kothmann. 1997. Structure and causes of vegetation change in state and transition models. Journal of Range Management 50: 399-408. Society for Range Management Task Group on Unity in Concepts and Terminology Committee. 1995. Evaluating rangeland sustainability. Rangelands 17: 85-92. Stoddart, L.A. 1941. The Palouse grassland association in Northern Utah. Ecology 22: 158-163. Stoddart, L.A., A.D. Smith, and T.W. Box. 1975. Range management. McGraw Hill, New York, NY. Suttie, J.M., S.G. Reynolds, and C. Batello. 2005. Grasslands of the world. Food and Agricultural Organization of the United Nations, Rome. Tisdale, E.W. 1947. The grasslands of the southern interior of British Columbia. Ecology 28:  20  346-382. Tisdale, J.M., and J.M. Oades. 1982. Organic matter and water-stable aggregates in soils. J. of Soil Sci. 33:141—163.  Trlica, M.J. and L.R. Rittenhouse. 1993. Grazing and plant performance. Ecological Applications 3: 21-23. Vallentine, J.F. 2001. Grazing management. Academic Press. San Francisco, CA. Vallentine, J.F. 1989. Range development and improvements. Diego, CA.  rd 3  ed. Academic Press, San  Warren,S.D., M.B. Nevill, and W.H. Blackburn. 1986. Soil response to trampling under intensive rotation grazing. Soil Science Society of America Journal 50: 1336-1341. Wikeem, B.M., A. McLean, A. Bawtree, and D. Quinton. 1993. An overview of the forage resource and beef production on crown land in British Columbia. Canadian Journal of Animal Science 73: 779-794. Wikeem, B. and S. Wikeem. 2005. Grassland assessment manual for British Columbia: A prototype for the Fescue Grasslands of the Nicola Valley. Version 2. Grassland Conservation Council of B.C. Woodmansee, R.G. 1978. Additions and losses of nitrogen in grassland ecosystems. BioScience 28: 448-453.  21  2. Development of Health Indicators for Rough Fescue Grasslands of British 1 Columbia 2.1 Introduction The need for sustainability criteria and indicators was identified at the Earth Summit in Rio de Janeiro (1992), and subsequently in Montreal (2000) and Johannesburg (2002), but little progress has been made in developing soil and plant indicators for temperate forest and grassland ecosystems. Studies on development and application of criteria and indicators for forests and grasslands are often lacking, or have been done on a limited number of sites with relatively narrow ranges of climate and soil type (Working Group on Criteria and Indicators for Conservation and Sustainable Management of Temperate and Boreal Forests 1999; Page Dumroese et al. 2000). Development of sensitive and unambiguous indicators of rangeland health should start with identification of vegetation and soil properties that integrate important ecological processes, followed by examination of linkages among these selected vegetation and soil properties and rangeland health assessment indicators. With this method, one can refine the boundaries of the rangeland health categories and determine if confidence can be placed in existing rangeland health assessments. The plant community approach to assessing rangeland condition came into question in the 1990s (SRM Task Group 1995) and rangeland health assessments were proposed as replacements for the traditional system based on plant seral stage of the site. Rangeland health assessment methods use an holistic approach to determine the state of rangelands, and include assessments of ecological functions in addition to plant community information. Rangeland health has been defined by Adams et al. (2005) as “The ability of rangeland to perform certain key functions.. where all parts that make up the whole are present and working together,” while .  Busby (1994) gave another definition “Rangeland health is the degree to which the integrity of the soil and the ecological processes of rangeland ecosystems are sustained.” Most rangeland health assessment methods have been designed to be simpler than traditional methods of rangeland condition assessment, allowing wider use with minimal training. Visual rating of indicators is often used in favour of quantitative measures. Because of the reliance on visual rating, there are some concerns regarding the reliability of rangeland health ‘A version of this chapter will be submitted for publication. Authors: Lamagna, S.F., M. Krzic, R.F. Newman, B. Wallace. Development of health indicators for fescue grasslands of British Columbia.  22  assessments. In British Columbia (BC), two commonly used rangeland health assessment methods are the BC Ministry of Forests and Range upland assessment method (British Columbia Ministry of Forest 1997) and the Grasslands Conservation Council method for the fescue grasslands (Wikeem and Wikeem 2005). The National Research Council (1994) and Herrick eta!. (2002) outlined the following three principal criteria for determination of rangeland health (1) degree of soil stability and watershed function, (2) integrity of nutrient cycles and energy flows, and (3) presence of functioning recovery mechanisms. To address these criteria various soil and vegetation properties were evaluated in this study and they are listed below.  Under criteria no.1. (degree of soil stability and watershed function) we focused on: a) soil bulk density and mechanical resistance, indicators of soil compaction, hydrologic function, and soil pore stability, b) soil aggregate stability, describes the resistance of soil structure to external disturbance.  Under criteria no.2. (integrity of nutrient cycles and energy flows) we focused on: a) total soil C and N, representing total inventory of soil organic matter, b) soil polysaccharides, indicator of the most readily available source of energy for soil micro-organisms and one of the factors that enhances aggregate stability, c) living above-ground biomass, an estimate of site productivity (energy capture), d) dead above-ground biomass, an indicator of nutrient cycling.  Under criteria no.3 (presence of functioning recovery mechanisms) we focused on: a) density of key late seral grasses, an indication of the rate of recovery possible b) flowering culm density of key late seral grasses, an indication of reproductive capacity to determine plant recruitment potential. Following a review by literature, the properties listed above were selected because they can be obtained using sound, well-established methods and/or they assess key processes known to be affected by grazing. To our knowledge, a limited number of studies have evaluated the linkages among indicators of rangeland health assessment and quantitative measures of  23  vegetation and soil properties (Miller 2008). Examining the relationships among these properties and rangeland health assessment indicators will improve effectiveness of health assessment systems used by rangeland managers to monitor the impacts of their practices. The objective of this study was to quantify the relationships among soil/vegetation properties known to be affected by grazing to easily-assessed indicators, used in the existing health assessment systems. The hypothesis was that soil and vegetation properties, known to be affected by grazing, are related to easily-assessed indicator variables used in health assessment systems.  2.2 Materials and Methods 2.2.1 Site Description The study was carried out on 20 treatment units consisting of ungrazed exciosures (between 0.5 and 1.0 ha in size) and adjacent grazed areas (Table 2.1). Treatment units were located at nine grassland sites within the Interior Douglas-fir biogeoclimatic zone in the southern interior of BC. All treatment units were characterized by open grassland with a potential natural plant community dominated by rough fescue (Festuca campestris Rydb.). At least one corresponding grazed treatment unit was selected at each site. The basic layout consisted of five 30-rn long transects established within each treatment unit. Transects were parallel to each other and spaced 5 m apart. Plant species composition was assessed along the transects while soil sampling occurred midway between transects to avoid damaging plants used for long-term assessment. Aboveground biomass sampling occurred adjacent to the plant assessment area. 2.2.2 Sampling and Analyses 2.2.2.1 Soil Sampling All soil samples were collected between May 8 to 15, 2007. Intact soil samples for bulk density determination (Blake and Hartge 1986) were taken with a drop-hammer sampler (with a core of 7.5 x 75 cm) from 0-7.5 cm depth. One bulk density sample was taken per transect within each treatment unit (i.e., five samples were collected per treatment unit). Soil samples for bulk density determination were dried at 105°C for 48 hours in a forced-air oven. Coarse fragments (diameter >2 mm) within the sample were  24  screened out and weighed. Volume of mineral coarse fragments were determined from dry mass and assumed to have a particle density of 2.65 Mg m . Soil bulk density was calculated as the 3 mass of dry, coarse fragment free mineral soil per volume of field-moist soil, where volume was also calculated on a coarse fragment free basis. Soil mechanical resistance (Bradford 1986) was measured to 4.5 cm depth. Measurements were recorded at depth intervals of 1.5 cm, using a hand-pushed 13-mm diameter cone  (300)  penetrometer with data logger (Agridry Rimik PTY Ltd., Toowoomba, QLD,  Australia). Six measurements were recorded at random locations along each of the five transects totalling 30 measurements per treatment unit. The soil aggregate stability was determined using a variation of the wet sieving method (Nimmo and Perkins 2002). Five samples were taken at 0-7.5 cm depth at random locations along each transect and combined to make one composite sample per transect. Field moist samples were hand-sieved using a 6-mm sieve and collected on a 2-mm sieve. The pre-sieved 2-6 mm moist sample (of about 10 g) was placed on the top of a nest of sieves with openings of 2, 1, and 0.25 mm and wetted in a humidifier for about 60 minutes to minimize disruption caused by air trapping. This was done immediately before wet sieving. Wet sieving was performed for 10 minutes in a motor-driven mechanical device with a vertical stroke of 2.5 cm at a rate of 30 strokes per minute. The motion of the system had both a vertical stroke and an oscillating action through an angle of 30°. After the sieves were removed from the water the proportion of material retained on each sieve was oven dried at 105°C for 24 hours, weighed, and expressed as a percentage of the total soil. The results for aggregate stability were expressed as the mean weight diameter (MWD), which represents the sum of the mean diameter of each size fraction (Di) and the proportion of the sample weight occurring in the corresponding size fraction (Wi). The summation was carried out over all four size-fractions, including the one that passed the 0.25-mm sieve (MWD  =  WD). Corrections were made for non-aggregate particles retained  on each sieve to avoid biased interpretations of water stable aggregates. Non-aggregate particles were subtracted from the mass of aggregates on each sieve by crushing all soil aggregates in a mortar and pestle then passing the material through the sieve and weighing the material that remained on the sieve.  25  Soil samples for total C and N were taken at 0-7.5 and 7.5-15 cm depths. Five samples were taken per each of the five transects and composited. Samples were air-dried and ground to pass a 2.0-mm sieve. Soil total C (Nelson and Somers 2002) and total N (McGill and Figueirdeo 1993) were determined by elemental analysis using an automated analyzer (LECO CHN-600 Elemental Analyzer). Five soil samples per transect were taken at 0-7.5 cm depth and composited for soil polysaccharide and pH analyses. Soil polysaccharides were determined by the phenol-sulfuric acid method of Dubois et al. (1956) as modified by Doutre et al. (1978). Soil pH was determined on a 1:1 (v/v) soil to distilled water saturated paste and 1:1 (v/v) soil to 0.01 M CaCl 2 solution (Hendershot et al. 1993). The percentage of ground surface composed of exposed mineral soil was assessed within 0.5 x 0.2 m quadrats using a Daubenmire frame. Ten assessments were completed at random locations along each of the 5 soil transects. Soil particle size distribution (i.e., texture) was determined by the hydrometer method (Gee and Or, 2002). 2.2.2.2 Plant Sampling All vegetation measurements and sampling were completed in August 2007 except for vascular plant cover and microbiotic crust, which were sampled in a previous study during June 2006. Living above-ground biomass was estimated from peak annual standing crop of grasses and forbs. One-half m 2 areas were clipped to ground level and the plant litter was separated from current year’s growth of forbs and grasses. Five plots were randomly located within each treatment unit. Living plant material was sorted by species group, stored in paper bags and airdried to minimize decomposition. Standing litter was retained and stored separately. Samples were oven-dried at 70°C to a constant weight, and weighed to the nearest 0.1 g. Living above ground biomass cannot be properly determined if any current annual growth is removed by grazing animals. Consequently, portable cages (1 xl xl m) were used to exclude grazing by domestic and wild ungulates on unfenced treatment units. Cages consisted of 1 m 2 meshed wire panels designed to permit light and precipitation in, but of a small enough mesh to prevent grazing.  26  Fallen litter (detached above-ground dead plant material not incorporated into mineral soil) was collected from within the plots used to determine living above-ground biomass (0.5 m 2 areas) and combined with standing litter samples. Litter samples were processed in the same manner as living material. The density of key bunchgrasses, rough fescue (Festuca campestris Rydb.), bluebunch wheatgrass (Pseudoroegneria spicata (Pursh) A. Love), and Idaho fescue (Festuca idahoensis Elmer) were determined within 1 m plots. Five plots were randomly located within each treatment unit. The number of flowering cuims of key bunchgrasses (i.e., rough fescue, bluebunch wheatgrass and Idaho fescue) were counted within 0.5 m plots. The abundance of flowering 2 cuims is related to reproductive capacity and plant recruitment potential and thus is an indication of the presence of functioning recovery mechanisms. Plant species composition and canopy cover was determined using a modified canopy coverage method (Daubenmire 1959) by assessing the cover of each plant species to the nearest percentage. Vascular plant species were identified to genus, species, and sub-species using nomenclature provided by Douglas et al. (1998). Ten canopy cover assessments were completed in June 2006 at random locations along each of the five transects. The percentage of ground surface covered by microbiotic crust (including lichen, algal crust, bryophytes) was assessed within 0.5 x 0.2 m plots using Daubenmire frames. Lichens and bryophytes were categorized by life-form. Ten assessments were done at random locations along each of the five transects in June 2006. 2.2.2.3. Grazing This study used the existing plant community seral stage as an integrating index of the history of grazing rather than attempt to quantify the grazing level using the prescribed use of the pasture over many years. The effects of grazing on plant succession have been known for almost 100 years (Sampson 1919). Improper cattle management can lead to retrogression of plant communities, while rest from grazing can result in recovery of plant communities (Dyksterhuis 1949). Secondary plant succession in the rough fescue plant community of BC is relatively well known (McLean and Marchand 1968; Lloyd et al. 2005). Newer concepts such as the “state and transition model” have proposed that multiple stable equilibrium communities and unstable  27  transition communities may exist (Laycock 1991; Westoby et a!. 1989). The significance of possible multiple stable equilibrium communities and unstable transition communities were considered during result interpretation. These developmental states were determined by comparing present-day plant communities with historical data from nearby rangeland reference areas. Non-equilibrium models are not relevant in rough fescue plant communities. Reference areas are valuable for understanding and predicting vegetative changes in plant communities (Holzman and Isaacs 1999). A rangeland reference area is: “an area set aside which illustrated or typifies virgin conditions of forest and range growth (or) other conditions that have special unique characteristics of scientific interest and importance from a range resource standpoint, to be retained primarily for the purpose of science, research, and education” (Laycock 1975). 2.2.3 Statistical Analysis Regression was used to determine if rangeland health assessment indicators were related to soil and vegetation properties known to be affected by grazing. Treatment units were subsampled from 3 to 50 times, depending on the parameter measured, to provide a better estimate of the mean. Simple linear and second degree polynomial models were tested.  2.3 Results and Discussion A total of 25 potential indicators (Appendices 1 and 2) were evaluated for their ability to predict rough fescue grassland health as defined by the National Research Council’s (1994) three principal criteria but only the relationships significant at P <0.05 are presented here (Table 2.2). 2.3.1 Principle Criteria no. 1 Degree of Soil Stability and Watershed Function -  Three properties were used to assess the degree of soil stability and watershed function at the grassland sites  —  soil bulk density, mechanical resistance, and aggregate stability. The first  two are measures of soil compaction, while aggregate stability is often used to characterize soil susceptibility to erosion, crust formation, and hard setting (Angers and Mehuys 1993). Soil bulk density had a relatively strong, negative relationship (R =0.47) with percent 2 litter cover, i.e., soil bulk density was reduced with increasing litter cover (Fig. 2.01). Low bulk density is desirable because it implies low soil compaction and consequently good water infiltration/aeration creating a favourable environment for root growth. Percent litter cover is an  28  easily measured parameter, which can be determined visually. The amount of litter cover is a direct assessment of the accumulation of one to three year-old dead plant material at the site. Litter cover increases with increases in the biomass of living plants. The “stubble height” of vegetation remaining after cattle grazing is directly related to the amount of litter at a site. Generally, increased grazing leads to a lower stubble height and less litter. Decomposing litter is a potential input for soil organic matter accumulation in grasslands, although its contribution is secondary to that of decomposing grass roots. Soil organic matter is inversely related to soil bulk density (Manrique and Jones 1991), which explains why soil bulk density decreased as the amount of litter on a site increased (Figure 2.01). Soil bulk density had a relatively strong positive relationship (R =0.46) with exposed 2 mineral soil (Fig. 2.02). Exposed mineral soil is a health assessment indicator that can be visually determined with greater repeatability than percent litter cover. This is because mineral soil exists on one stratum only (ground level) as compared to litter, which is found detached at the soil surface as well as attached to plants in the plant canopy. Increased mineral soil exposure is generally an indication of a loss of plant cover resulting from reduced plant vigour, reduced plant density due to plant mortality, and/or reduced plant size due to a change in plant species composition. Burrowing activity by rodents can also enhance soil exposure (Evans 1997). These scenarios imply that a loss of plant cover leads to a reduction in organic matter input to the soil, thus leading to an increase in soil bulk density. Soil bulk density had positive relationships with percentage cover of both Junegrass (Koeleria macrantha (Ledeb.) J.A. Schult. f) and Sandberg’s bluegrass (Poa secunda J. Presi). Bulk density increased with increasing Junegrass cover until 15% cover was reached (Fig. 2.03) and there was no relationship with bulk density at Junegrass cover above 15%, although more data points are required to confirm this. In contrast to that found with Junegrass, soil bulk density increased almost linearly across the range of Sandberg’s bluegrass cover (Fig. 2.04), but Sandberg’s bluegrass was less suitable as an indicator because it only occurred at about half of the study sites (Fig. 2.04). Both Junegrass and Sandberg’s bluegrass are plant species that increase with grazing in the rough fescue grasslands of BC (McLean and Marchand 1968). In other words, they are species that occur at lower seral stages. Sandberg’s bluegrass is usually considered an early seral species in the rough fescue grasslands, while Junegrass is considered an intermediate seral  29  species. The increase in soil bulk density on sites where the cover of these two plant species is high cannot be directly attributed to the plants themselves (i.e., Sandberg’s bluegrass roots do not compact the soil). Trampling by grazing cattle most likely leads to soil compaction concurrently changing plant species composition. The presence of Sandberg’s bluegrass is actually an indicator of previous improper grazing practices and inadequate periods of rest. This is particularly pertinent when differences in Sandberg’s bluegrass cover occur between an exclosure and an adjacent cattle-grazed area. Plant species have a differing ability to ameliorate soil compaction, due to different root structure and rooting depths. For example, roots of Sandberg’s bluegrass are distributed at a shallower depth than roots of many other larger bunchgrass species (Link et al. 1990). Therefore, greater root biomass below larger bunchgrasses (such as rough fescue) will result in greater resistance to soil compaction over time compared to Sandberg’s bluegrass. Soil mechanical resistance at 4.5 cm depth was relatively strongly related to many of the same indicators as soil bulk density, and had the same type of relationship with those indicators. Soil mechanical resistance was reduced with increasing percent litter cover (Fig. 2.05). Reduction of soil mechanical resistance is desirable since it implies easier root growth. For example, it has been commonly cited that root growth is reduced above 2000 kPa, as measured by a flat-tipped penetrometer (Gerard et al. 1982, Busscher and Sojka 1987), which corresponds to about 2500 kPa for the 13-mm cone tip (30’) penetrometer (Busscher et al. 1986). Soil mechanical resistance increased with increasing exposed mineral soil (Fig. 2.06) and increasing cover of Sandberg’s bluegrass (Fig. 2.07). A weaker, but nonetheless interesting relationship (R 2 =  0.28) existed between soil mechanical resistance and percent cover of rough fescue. Lower soil  mechanical resistance was related to greater cover of rough fescue (Fig. 2.08). With a greater amount of rough fescue cover, it is not exposed to the elements and creates a barrier between the soil and cattles’ hooves. Rough fescue decreases the impact hooves will have on the compaction of soil. Soil mechanical resistance appeared insensitive to cover of rough fescue when rough fescue was not frequent at the site as indicated by the wide variation of soil mechanical resistance values at rough fescue cover values below 25% (Fig. 2.08). This trend may be due to different plant species replacing rough fescue during periods of retrogression. If the replacement species was bluebunch wheatgrass, soil mechanical resistance would be unchanged. If rough fescue was replaced by less desirable plant species (e.g., Kentucky bluegrass [Poapratensis L.j  30  or common dandelion [Taraxacum officinale G.H. Weber ex Wiggers]), mechanical resistance would increase. Aggregate stability is evaluated by determining the size distribution of aggregates and mean weigh diameter (MWD). The aggregate size distribution determines (i) their susceptibility to movement (erosion) by wind and water and (ii) size of pores, which in turn affects the movement and distribution of water and air in the soil. When unstable aggregates are exposed to external stress (e.g., grazing, tillage) they easily break apart into small aggregates and primary mineral particles (i.e., clay, silt, sand) that then clog soil pores used for water drainage and gas exchange through the soil profile. The proportion of water stable aggregates in the 1—2 mm size class was the size fraction that was best predicted by percent litter cover, Sandberg’s bluegrass cover, and percent of exposed mineral soil (Figs. 2.09 2.11), as compared to all other aggregate size fractions. -  Percent of litter cover had a positive relationship with the proportion of aggregates in the 1—2 mm size class (Fig. 2.09). Greater litter cover provided a better protection of aggregates at the soil surface and most likely provided better conditions for enhanced microbial activity. Both impacts would in turn enhance aggregate stability. The percent cover of Sandberg’s bluegrass was inversely related to the proportion of aggregates in the 1—2 mm size class (Fig. 2.10). Although the majority of points were close to 0% Sandberg’s bluegrass cover, the remaining sites showed a consistent negative relationship with increasing aggregates in the 1-2 mm size class. Sandberg’s bluegrass is an early seral species and would generally establish on sites that have had high grazing pressures. This relationship further emphasizes the need to investigate the usefulness of specific plant species, such as Sandberg’s bluegrass, as an indicator species for grassland health. The proportion of aggregates in the 1—2 mm size class was inversely related to exposed mineral soil (Fig. 2.11). This was in agreement with the relationship between litter cover and stable aggregates in this size class (Fig. 2.09). Sites with a high degree of exposed mineral soil generally have a lower soil water content. Evaporation is encouraged from the lack of an insulating layer that plants and litter provide at the soil surface (Molinar et al. 2001). In grassland ecosystems that are facing limiting water supply, an increase in soil water content usually leads to increases in soil biological activity and plant vigour, which encourages the formation and stabilization of soil aggregates (Tisdale and Oades 1982).  31  A positive relationship was observed between 2-6 mm size aggregates and cover of timber milkvetch (Astragalus miser Dougl. ex. Hook) (Fig. 2.12). The MWD exhibited a similar trend (Fig. 2.24) as the fraction of soil in the largest, 2-6 mm aggregate size class. This was not surprising since aggregates from the 2-6 mm size class have the greatest mass and mean diameter as compared to the other aggregate size classes, and therefore strongly influence MWD, which is expressed as the sum of the product of mean aggregate size and mass of soil retained on each sieve. Timber milkvetch is a leguminous plant species that fixes N from the atmosphere. Sites with high cover of this species may be able to support a larger soil microbial population (Biederbeck et a!. 2005), which would in turn lead to greater aggregate stability. It is possible that this enhanced N availability and microbial activity due to a greater presence of timber millkvetch obscured negative grazing impacts on aggregate stability that could be expected on sites with greater cover of this plant species. This relationship requires further study because it was not an expected outcome. Specific plant species are not necessarily the cause of measured effects in soil properties, such as aggregate stability, but species such as timber milkvetch may indicate certain site characteristics that favour the formation and stabilization of soil aggregates. Generally, greater percentage of timber milkvetch cover indicates lower ecological status (e.g., lower seral stage) of a grassland. It has been observed that timber milkvetch initially decreases with grazing, and as grazing pressure intensifies, the cover of this species increases (Majak eta!. 1996). 2.3.2 Principle Criteria no. 2 Integrity of Nutrient Cycles and Energy Flows -  Four properties were used to assess the integrity of nutrient cycles and energy flow at the rough fescue grassland sites  —  plant biomass (both total aboveground living biomass and litter  biomass), total soil C, total soil N, and soil polysaccharides. The total aboveground living biomass (estimated from peak annual standing crop of grasses and forbs) had a moderately strong (R =0.41) positive relationship with rough fescue 2 cover (Fig. 2.27). This is to be expected since rough fescue is the dominant plant species at the grassland sites studied, often contributing over 90% of the total aboveground living biomass. Rough fescue cover predicted total aboveground living biomass fairly consistently across the range of cover values, although considerable variance does exist. No other indicator out of 25 that were evaluated reliably predicted total aboveground living biomass.  32  Litter biomass increased with increasing rough fescue cover up to about 50% after which the relationship ceased to change (Fig. 2.28). There was a clearly separate group of low litter biomass values when rough fescue cover was 25% or lower. This grouping may be useful for establishing a management threshold (rough fescue cover> 25% vs. rough fescue cover < 25%), although further sampling will be required to verify this. Average litter biomass was 737 kg ha 1 for the low rough fescue cover group and 3837 kg ha’ for the high rough fescue cover group. It is apparent that rough fescue plants are the primary contributor to litter biomass on these study sites. Three parameters were found to have negative relationships to soil C. Soil C at the 0— 7.5 cm depth was reduced with increasing mineral soil exposure (Fig. 2.13). Aboveground and associated belowground plant biomass are major contributors of soil organic matter in grassland ecosystems (Coupland and Brayshaw 1953). Hence, it is not surprising that a negative relationship was found between the soil C and exposed mineral soil. Cover percentage of Sandberg’s bluegrass had a similar relationship to soil C as did mineral soil exposure, where increase in Sandberg’s bluegrass cover resulted in a reduction of soil C at the 0— 7.5 cm depth (Fig. 2.14). Sandberg’s bluegrass is an early seral species in the rough fescue grasslands, and is often associated with plant communities with low plant cover/biomass. Sandberg’s bluegrass does not rob the soil of C itself, but its presence indicates the loss of more productive plant species such as rough fescue. Rough fescue is a major contributor to litter on sites that are in mid to late seral stages (Fig. 2.27) and is known to have extensive root systems, contributing to high inputs of organic matter to the soil. Cover percentage of Junegrass was also negatively associated with soil C (Fig. 2.16); however, the lower depth (7.5  —  15 cm) sampled was more sensitive than the upper depth (0  —  7.5 cm). Further study may reveal more information that would help to explain why the relationship with soil C at the 7.5  —  15 cm depth was more sensitive to Junegrass cover  (Coupland and Johnson 1965). Litter cover was positively correlated with soil C from the 7.5 The increase in soil C at the lower depth (7.5  —  —  15 cm depth (Fig. 2.15).  15 cm) in conjunction with increasing litter may  be partly explained by the decomposition of roots, which is known to be the largest organic matter input responsible for increasing C storage in grassland soils (Sousanna et a!. 2004). An  33  increase in litter cover generally implies that inputs of both above and below ground biomass have increased which would have resulted in a greater soil C. Soil N had similar relationships to cover percentage of Sandberg’s bluegrass (Fig. 2.17), litter cover (Fig. 2.18), exposed mineral soil (Fig. 2.19), and Junegrass (Fig. 2.20) as did soil C. This was not surprising since the ratios of C to N in soils across all sites remains fairly constant with values between 11 and 13. This finding is in agreement with the fact that C to N ratio is relatively consistent (i.e., in the 10 to 14 range) for different grassland soils under a wide range of management and environmental conditions (Jenkinson 1988). Total soil C and N represent the total inventory of soil organic matter, which is considered to be a key attribute of soil quality and environmental quality (Gregorich et at. 1994; Carter 2002). Soil organic matter is involved in many soil chemical, biological, and physical properties such as susceptibility to compaction, friability, soil-water holding capacity, aggregation, potential water infiltration, nutrient supply, and susceptibility to erosion. Assessment of soil organic matter is a valuable step towards identifying the overall quality of a soil and the whole ecosystem (Christiansen and Johnston 1997). Soil organic matter consists of a range of compounds from very stable to biologically active, including readily decomposable materials, litter and root biomass, and dead and living soil organisms. As mentioned above, soil total C and N provide a measurement of a soil’s total inventory of organic matter, while an array of properties. The light fraction of the soil organic matter, mineralizable N, microbial biomass, carbohydrates, soil enzymes reflects the part of organic matter that is relatively easily decomposed (Gregorich et al. 1994). Carbohydrates in general, and polysaccharides in particular, contribute to soil and grassland health through their role in the stabilization of soil structure and the provision of a readily available source of energy for soil organisms. Soil polysaccharides are compounds of a wide range of monosaccharides in both simple and complex molecular structures that comprise a small but nonetheless important part of soil organic matter (Haynes and Swift 1990; Tisdale and Oades 1982). Soil polysaccharides were negatively related to the cover percentages of Junegrass and common dandelion (Figs. 2.21 and 2.22), which is consistent with the idea that these plant species indicate a disturbed site with low productivity. Consequently, soil biological activity during the decomposition process, as indicated by polysaccharide content (Flaynes and Swift 1990; Tisdale and Oades 1982), would be low on sites with high dominance of these two plant species due to the low organic matter inputs  34  to the soil. A positive relationship between litter biomass and polysaccharide content (Fig. 2.23) further supported the idea that greater above-ground biomass means greater organic matter input that stimulates soil biological activity. Increase in the litter biomass resulted in an increase of polysaccharides, but only up to about 3000 kg ha 1 and additional increases in litter biomass did not result in an increase of soil polysaccharides. It appears that polysaccharides reached a maximum or equilibrium level, and that further litter inputs do not contribute in a positive way. Plant litter helps conserve soil water by reducing soil temperature and evaporation, yet the same mechanism can delay plant growth and microbial activity in spring (Wilims et al. 1986). 2.3.3 Principle Criteria no. 3 Presence of Functioning Recovery Mechanisms -  Two properties were used to assess the presence of functioning recovery mechanisms  —  seed head density and bunchgrass density. Although the presence of seed heads alone does not ensure that viable seed will be produced, the density of seed heads provides an indication of the potential for sexual reproduction. No indicator was related to total bunchgrass seed heads. Rough fescue cover was found to be strongly related to rough fescue seed heads and the increased presence of rough fescue resulted in increased density of rough fescue seed heads (Fig. 2.27). The presence of rough fescue; however, did not necessarily mean that seed heads were produced. For example, an ungrazed exclosure with 76% cover of rough fescue did not produce any seed heads in 2007. The same exclosure produced 44 seed heads per m 2 in 2006 (data not shown). Seed production of rough fescue is known to be erratic and the factors responsible for this behaviour are not well understood (Johnston and MacDonald 1967; Stout et al. 1981). The density of key bunchgrasses (rough fescue, bluebunch wheatgrass, and Idaho fescue) was only related to one factor, rough fescue percent cover. Increasing density of key bunchgrasses was strongly related (R =0.76) to increasing rough fescue cover (Fig. 2.28) because 2 rough fescue was, by far, the most common bunchgrass on the 20 treatment units sampled. On average, 85% of the bunchgrass plants sampled were rough fescue. Bluebunch wheatgrass and Idaho fescue formed only 13 and 2% of the total density, respectively. It is not surprising that rough fescue cover will predict rough fescue density. The relationship is, nonetheless, important. Rough fescue cover provides a strong measure of the ability for the recovery of key  35  bunchgrasses. This relationship is even more critical given the fact that rough fescue seed production is erratic. 2.3.5 Examination of Linkages Among Rangeland Health Indicators and Quantitative Vegetation and Soil Properties Eight indicators were selected from the list of 25 potential indicators for further detailed examination (Table 2.3) based on the strength of their relationships with vegetation and/or soil properties known to be affected by grazing. Five of the eight indicators (i.e., percent litter cover, percent exposed mineral soil, and percent cover of three native grass species  —  Sandberg’s  bluegrass, Junegrass, and rough fescue) were found to have relatively strong relationships with multiple vegetation and soil properties (Table 2.3) and therefore received top ranking (Rank=l). No indicator by itself was useful for predicting all three principle criteria; however, the selected five indicators were each related to two of the criteria (Table 2.3). Percent litter cover and percent exposed mineral soil were consistently related to Principle Criteria no. 1 (degree of soil stability and watershed function) and Principle Criteria no. 2 (integrity of nutrient cycles and energy flows). Percent litter cover always had an inverse relationship to percent exposed mineral soil for the same health criteria. Litter was always associated with better ecosystem health, while exposed soil was always associated with poor health. Because of this, it may not be necessary to include both indicators when determining rangeland health. Percent cover of rough fescue stands apart from the other indicators because it was the only reliable predictor of Principle Criteria no. 3 (the presence of functioning recovery mechanisms). The presence of functioning recovery mechanisms, as defined by key bunchgrass plant density and bunchgrass seed head density, requires the assessment of rough fescue cover. It should be noted; however, that the measurement of bunchgrass plant density and bunchgrass seed head density are not much more difficult to determine directly than percent rough fescue cover. Rough fescue was the only reliable indicator of aboveground and litter biomass (Principle Criteria no. 2). Junegrass and Sandberg’s bluegrass had negative associations with rangeland health, because they increase with increasing grazing intensity. These two grass species become more common at earlier seral stages. Although they both often showed similar relationships to the  36  same set of Principle Criteria, they appear to contribute in different ways. Sandberg s bluegrass ‘  is common on sites in poor health, but did not occur on many of the sites in better health. Junegrass tended to be more evenly distributed across the sites. Junegrass may prove to be a useful early indicator of poor grazing practices, while Sandberg’s bluegrass may be useful in differentiating among the poorest sites. Further examination of the five top indicators using correlation analysis revealed that percent Sandberg’s bluegrass was closely associated with percent exposed mineral soil (r = 0.87) (Table 2.4). This close association means that they are likely sensitive to the same health parameters; therefore, both are not needed. In fact, they were associated with the same dependent variable four out of five times (Table 2.2). Percent litter cover also had a strong, but negative association with percent exposed mineral soil (r = -0.82), as previously observed (Table 2.4), indicating possible redundancy. The use of either exposed mineral soil or Sandberg’s bluegrass cover, but not both, is recommended. Similarly, the use of either litter cover or exposed mineral soil, but not both, is recommended.  2.4 Conclusions The relationships among vegetation/soil properties and rough fescue rangeland health indicators were quantified for the southern interior of BC. Twenty five potential indicators were evaluated for their ability to predict rough fescue rangeland health as defined by National Research Council’s three principal criteria. The top five rangeland health indicators based on strength of relationship (R ) with vegetation and soil properties (known to be affected by 2 grazing) were: percent exposed mineral soil, percent litter cover, percent Sandberg’s bluegrass cover, percent Junegrass cover, and percent rough fescue cover. Two of these five indicators (i.e., percent litter cover and percent Sandberg’s bluegrass cover) showed redundant function with percent exposed mineral soil suggesting that they do not contribute unique information to an assessment of health. Percent rough fescue cover is an essential indicator of the presence of functioning recovery mechanisms. There were no substitute indicators found that can be used to assess this particular health category. Percent exposed mineral soil is a sensitive indicator of the degree of soil stability and watershed function, as well as an indicator of the integrity of nutrient cycles and  37  energy flows in rough fescue grasslands. Percent Junegrass cover is not as sensitive an indicator as percent exposed mineral soil, but has general overall strength with many health measures. These observations show that there is a consistency among easily-assessed rangeland health indicators and quantitative vegetation and soil properties. With these findings, a set of indicators can be established throughout the rangeland community that ranchers and government officials can use to assess the health of the rangeland in the southern interior of British Columbia. These findings may also be useful for rough fescue rangeland found outside of British Columbia.  38  2.5 References Adams, B.W., G. Ehiert, C. Stone, D. Lawrence, M.Alexander, M. Willoughby, C.Hincz, D. Moisey, and A. Bogen. 2005 Rangeland Health Assessment for Grassland, Forest and Tame Pasture. Alberta Sustainable Resource Development. Public Lands Division. Edmonton, AB. Angers, D.A., and J. Caron. 1998. Plant-induced changes in soil structure: processes and feedbacks. Biogeochemistry 42:55—72. Angers, D.A. and G.R. Mehuys. 1993. Aggregate stability to water. p. 65 1-658. In M.R. Carter (ed.) Soil sampling and methods of analysis. Lewis Publisher, Boca Raton, Florida. Biederbeck, V.0., R.P. Zentner, and C.A. Campbell. 2005. Soil microbial populations and activities as influenced by legume green fallow in a semiarid climate. Soil Biol. Biochem. 37:1775—1784. Blake, G.R. and K.H. Hartge. 1986. Bulk density. p. 363-375. In A. Klute (ed.), Methods of soil analysis. Part 1. Physical and mineralogical methods. Agron. 9 Amer. Soc. Agron., Madison, Wisc. Bradford, J.M. 1986. Penetrability. p. 436-478. In A. Klute (ed.) Methods of soil analysis. Part 1. Physical and mineralogical methods. Agron. 9 Amer. Soc. Agron., Madison, Wisc. British Columbia Ministry of Forests and Range. 1997. Range and riparian assessment reference manual. Ministry of Forests, Forest Practices Branch, Range section, Victoria, B.C. Busby, F.E. 1994. Rangeland health: new methods to classify, inventory, and monitor rangelands. NRC, Washington, D.C. Busscher, W.J. and R.E. Sojka. 1987. Enhancement of subsoiling effect on soil strength by conservation tillage. Trans. ASAE 30:888-892. Busscher, W.J., R.E. Sojka, and C.W. Doty. 1986. Residual effects of tillage on coastal plain soil strength. Soil Sci. 141:144-148. Carter, M.R. 2002. Soil quality for sustainable land management: organic matter and aggregation interactions that maintain soil functions. Agron. J. 94: 38-47. Christensen, B.T. and A.E. Johnston. 1997. Soil organic matter and soil quality lessons learned from long-term experiments at Askov and Rothamsted. P. 399-430. In E.G. Gregorich and M.R. Carter (eds.) Soil quality for crop production and ecosystem health. Elsevier, Amsterdam. —  Coupland, R.T. and T.C. Brayshaw. 1953. The fescue grassland in Saskatchewan. Ecology 34: 386-405.  39  Coupland, R.T. and R.E. Johnson. 1965. Rooting characteristics of native grassland species in Saskatchewan. Journal of Ecology 53: 475-507. Daubenmire, R. 1959. A canopy-coverage method of vegetation analysis. Northwest Sci. 33 :43-65. Douglas, G.W., Straley, G.B., Meidinger, D., and Pojar, J. (Eds.) 1998-200 1. Illustrated flora of British Columbia. B.C. Ministry of Environment, Lands and Parks and B.C. Ministry of Forests. Victoria, B.C. Doutre, D.A., G.W. Hay, A. Hood, and G.W. VanLoon. 1978. Spectrophotometric methods to determine carbohydrates in soil. Soil Biol. Biochem. 10:457-462. Dubois, M., K.A. Gilles, J.K. Hamilton, P.A. Rebers, and F. Smith. 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28:350-356.  Dyksterhuis, E.J. 1949. Condition and management of rangeland based on quantitative ecology. Journal of Range Management 2: 104-115. Evans, R. 1997. Soil erosion in the UK initiated by grazing animals. Applied Geography 17:127-14 1. Gee, G. and D. Or. 2002. Particle-size analysis. p. 255-293. In J.H. Dane and G.C. Topp (eds.) Methods of soil analysis. Part 4. Physical methods. SSSA Book Series no. 5. SSSA, Madison, WI. Gerard, C.J., P. Sexton, and G. Shaw. 1982. Physical factors influencing soil strength and root growth. Agron. J. 74:875-879. Gregorich, E.G., M.R. Carter, D.A. Angers, C.M. Monreal, B.H. Ellert. 1994. Towards a minimum data set to assess soil organic matter quality in agricultural soils. Can J. Soil Sci. 74: 367-385. Haynes, R.J. and R.S. Swift. 1990. Stability of soil aggregates in relation to organic constituents and soil water content. Eur. J. Soil Science. 41: 73-83. Hendershot, W.H., H. Lalande and M. Duquette. 1993. Soil reaction and exchangeable acidity. p. 141-145. In M.R. Carter, ed. Soil Sampling and Methods of Analysis. Canadian Society of Soil Science, Lewis Publishers, Boca Raton, FL. Holzman, B.A. and J. Isaacs. 1999. California rangeland reference area (RRA) inventory and database. Rangelands 21: 22-26.  40  Jenkinson, D.S. 1988. Soil organic matter and its dynamics. p. 564-607. In A. Wild (ed.) th Russell’s soil conditions and plant growth. 1 1 ed., Longman Group Ltd., UK. Johnston, A. and M. D. MacDonald. 1967. Floral initiation and seed production in Festuca scabrella Torn Can. J. Plant Sci. 47: 577-5 83. Laycock, W.A. 1975. Rangeland reference areas. Range Science Series #3. Society for Range Management. Denver, CO. Laycock, W.A. 1991. Stable states and thresholds of range condition on North American rangelands: a viewpoint. Journal of Range Management 44: 427-433. Link, S.O., G.W. Gee, and J.L. Downs. 1990. The effect of water stress on phenological and ecophysiological characteristics of cheatgrass and Sandberg’s bluegrass. Journal of Range Management 43: 506-513. Lloyd, D.K., K. Angove, G. Hope, and C. Thompson. 1990. A guide to site interpretation for the Kamloops Forest Region. B.C. Ministry of Forests, Land Management Report 23. Part 1, Res. Branch, Victoria, BC. Magurran, A.E. 1988. Ecological diversity and its measurement. Princeton University Press, Princeton, New Jersey. Majak, W., L. Stroesser, J.W. Hall, D.A. Quinton, and H.E. Douwes. 1996. Seasonal grazing of Columbia milkvetch by cattle on rangelands in British Columbia. Journal of Range Management 49: 223-227. Manrique, L.A., and C.A. Jones. 1991. Bulk-density of soils in relation to soil physical and chemical properties. Soil Sci. Soc. Am. J. 55:476-481. McLean, A., and L. Marchand. 1968. Grassland ranges in the southern interior of British Columbia. Can. Dept. Agric. Pub. 1319, Ottawa, Ont. McGill, W.B. and C.T. Figueiredo. 1993. Total nitrogen. p. 201-21 1 In M.R. Carter (ed), Soil sampling and methods of analysis. Can. Soc. Soil Sci., Lewis Publ., Boca Raton, Fla. Miller, M.E. 2008. Broad-scale assessment of rangeland health, Grand Staircase-Escalante National Monument, USA. Rangeland Ecology and Management 61: 249-262. Molinar, F., D. Galt, and J. Holechek. 2001. Managing for mulch. Rangelands 23:3-7. Nelson, D.W. and L.E. Sommers. 1996. Total carbon, organic carbon and organic matter. p. 961-1010 In D.L. Sparks et al. (eds.), Methods of soil analysis, Part 3, Chemical methods. No. 5 in SSSA Book Series, SSSA-ASA, Madison, Wise.  41  Nimmo, J.R. and K.S. Perkins. 2002. Aggregate stability and size distribution. p. 3 17-328. In A. J.H. Dane and G.C. Topp (ed.) Methods of soil analysis: Physical methods. Part 4. Number 5 in the Soil Science Society of America Book Series. Madison, Wisc. National Research Council. 1994. Rangeland health. New methods to classif’, inventory and monitor rangelands. NRC., Washington, D.C. Page-Dumroese, D., Jurgensen M., Elliot W., Rice T., Nesser J., Collins T. and Meurisse R. 2000. Soil quality standards and guidelines for forest sustainability in northwestern North America. For. Ecol. Manage. 138:445-462. Sampson, A.W. 1919. Plant succession in relation to range management. U.S. Department of Agriculture Bulletin No. 791. Washington, D.C. SAS Institute. 1988. SAS/STAT User’s Guide. Release 6.03 edition. SAS Institute Inc., Cary, NC. 1028 p. Soussana, J.F., P. Loiseau, N. Vuichard, E. Ceschia, J. Balesdent, T. Chevallier, and D. Arrouays. 2004. Carbon cycling and sequestration opportunities in temperate grasslands. Soil. Use. Manage. 20:219-230. Society for Range Management Task Group on Unity in Concepts and Terminology Committee. 1995. Evaluating rangeland sustainability. Rangelands 17: 85-92. Stout, D.G., A. McLean, and D.A. Quinton. 1981. Growth and phenological development of rough fescue in interior British Columbia. J. Range Manage. 34:455-459. Tisdale, J.M., and J.M. Oades. 1982. Organic matter and water-stable aggregates in soils. J. of Soil Sci. 33:141—163. Westoby, M., B. Walker, and I. Noy-Meir. 1989. Opportunisitc management for rangelands not at equilibrium. Journal of Range Management 42: 266-274. Wikeem, B. and S. Wikeem. 2005. Grassland assessment manual for British Columbia: A prototype for the Fescue Grasslands of the Nicola Valley. Version 2. Grassland Conservation Council of B.C. Willms, W.D., S. Smoliak, and A.W. Bailey. 1986. Herbage production following litter removal on Alberta native grasslands. J. Range Manage. 39:536—540. Working Group on Criteria and Indicators for Conservation and Sustainable Management of Temperate and Boreal Forests. 1999. Montreal Process.  42  Table 2.1. Location, site characteristics, and time of establishment of study sites in the southern  interior of British Columbia. Site  # 1 T Us  Date established  Latitude  Longitude  GooseLake Summit North Lac Du Bois Fertilizer Trial -Poa Lac Du Bois Fertilizer Trial Fescue Tunkwa Lake New Tunkwa Lake Old Drum Lake Agriculture Canada Weather Station Deep Lake Fertilizer Trial Fescue Microwave Repeater  2 2  1931 1968  5006 12” 5003 49”  12002537  2  1981  50°48’41”  2  1981  50°48’8”  1  1993  2 3  Elevation (m) 1160 1240  Slope  120026 1”  Aspect  BGC 2  (%)  (°)  5 5  125 180  IDFdkla IDFdkla  1000  15  81  IDFxh2  120°25’52”  947  10  107  IDFxh2  50° 35’ 56”  1200  51’ 56”  1150  4  25  IDFdkla  1963 1994  50°35’56” 50° 35’ 36”  120°51’56” 120° 40’ 26”  1200 1035  5 5  25 170  IDFdkla IDFdkla  2  1978  50° 47’ 12”  120° 26’ 56”  920  7  270  IDFxh2a  2  1981  50°47’35”  120°22’49”  900  15  90  IDFxh2a  2  —1975  50°4’27”  120°25’31”  1306  2  250  IDFdkla  1200  25’ 46”  -  -  ‘TU=Treatment units  BGC=biogeoclimatic classification according to Lloyd et al. 1990 2  43  Tabie 2.2. Summary of regression analyses that correspond to the three principal criteria for determination of rangeland health as outlined by National Research Council (1994) (n=20). Dependent Variable  Indicator Variable  P  2 R  0.005  0.47  0.005  0.46  Criteria no. 1 Degree of soil stability and watershed function -  (%)  Bulk density (Mg rn) 3  Litter cover  Bulk density (Mg rn) 3  Exposed mineral soil  Bulk density (Mg rn) 3  Koeleria macrantha cover (%)  0.001  0.55  Bulk density (Mg rn) 3 Mechanical resistance (kPa) at 4.5 cm soil depth Mechanical resistance (kPa) at 4.5 cm soil depth Mechanical resistance (kPa) at 4.5 cm soil depth  Poa secunda cover (%) Litter cover (%) Exposed mineral soil (%) Poa secunda cover (%)  0.009 0.017 0.023 0.004  0.42 0.38 0.36 0.48  Fraction of soil in I  2 mm aggregate class (kg kg-’)  Litter cover (%)  0.018  0.37  2 mm aggregate class (kg kg)  (%) Exposed mineral soil (%) Astragalus miser cover (%) Astragalus miser cover (%)  0.030  0.34  0.002  0.52  0.005 0.020  0.37 0.37  0.002  0.41  Litter biomass (kg ha-’)  (%) Festuca campestris cover (%)  0.004  0.48  Total C (0 7.5 cm depth) (%) Total C (0 7.5 cm depth) (%) Total C (7.5 15 cm depth) (%) Total C (7.5 15 cm depth) (%) Total N (0 7.5 cm depth) (%) Total N (7.5 15 cm depth) (%) Total N (7.5 15 cm depth) (%) Total N (7.5 15 cm depth) (%)  Exposed mineral soil (%) Poa secunda cover (%) Litter cover (%) Koeleria macrantha cover (%) Poa secunda cover (%) Litter cover (%) Exposed mineral soil (%) Koeleria macrantha cover (%)  0.005 0.018 0.004 0.034 0.018 0.003 0.014 0.035  0.37 0.28 0.47 0.49 0.28 0.50 0.40 0.49  Soil polysaccharides (tg mL’) Soil polysaccharides Qig mL-’)  Koeleria macrantha cover (%) Taraxacum officinale cover (%)  0.006 0.039  0.45 0.32  Soil polysaccharides (tg mL-’)  Litter biomass (kg ha-’)  0.022  0.36  Criteria no. 3 Presence of functioning recovery mechanisms ) 2 Festuca campestris flowering culm density (per m Bunchgrass plant density (>4 cm diam.) (per m ) 2  Festuca campestris cover (%) Festuca campestris cover (%)  0.034 >0.001  0.33 0.76  Fraction of soil in 1 Fraction of soil in I  -  -  -  2 mm aggregate class (kg kg-’)  Fraction of soil in 2 6mm aggregate class (kg kg-’) Soil aggregate mean weight diameter (mm) -  (%)  Poa secunda cover  Criteria no. 2 Interitv of nutrient cycles and energy flows -  Total aboveground living biomass (kg ha)  -  -  -  -  -  -  -  -  Festuca campestris cover  -  44  Table 2.3. Summary of indicators examined and their associations with the three principal criteria for determination of rangeland health as outlined by the National Research Council (1994) at 20 rough fescue grassland treatment units in the southern interior BC. Indicator description Exposed mineral soil (%) Koeleria macrant/ia cover (%) Litter cover (%) Poa secunda cover (%) Festuca campestris cover (%) Taraxacum officinale cover Astragalus miser cover (%)  (%)  Associated principle criteria 1 &2 1 &2 1 &2 1 &2 2&3  Indicato r Rank 1 1 1 1 1  2 1  2 2  Litter biomass (kg ha-’)  2  2  A ilium cernuum cover (%) Bryophytes cover (%) Crustose lichen cover (%) Festuca campestris biomass (kg ha-’)  NA NA NA NA  3 3 3  Festuca campestris cover (%) Festuca idahoensis small plant seed heads (per m ) 2  NA NA  Forbs biomass (kg ha-’)  NA  3  Other grass biomass (kg ha’)  NA  Poa pratensis cover (%) Pseudoroegneria spicata small plant density (per m ) 2  NA NA  3 3  Pseudoroegneria spicata small plant seed heads (per m ) 2  NA  Rosa acicularis cover (%) Sand(%) Silt (%) Soil pH at 0 7.5 cm depth Soil pH at 0 7.5 cm depth (CaC1 ) 2 Soil water content (%)  NA NA NA NA NA NA  -  -  3 3 3  3 3 3 3 3 3 3 3  45  Table 2.4. Correlation among the top five indicators: Sandberg’s bluegrass, percent exposed mineral soil, percent cover of Junegrass, percent litter, and percent cover of rough fescue.  Poa secunda cover (%)  Exposed mineral soil (%) Koeleria macrantha cover (%) Litter cover (%) Festuca campestris cover (%)  Poa secunda  Exposed mineral  Koeleria macrantha  Litter cover  Festuca campestri s cover  cover (%)  soil (%)  cover (%)  (%)  (%)  1 0.87 0.55 -0.65  -0.42  1 0.46 -0.82 -0.44  1 -0.45  -0.43  1 0.48  46  60  o o  •  Early Mid Late  Litter cover  70  (%)  80  2 y = -0.0001x 2 = 0.47** R n = 20 +  90  0.009x +  0.86  0.0  0.2  0.4  0 15  2 y = -0.00 1x 2 = 0.56*** R n = 20 +  0.03x +  0.73  5  Mid  Late  0  0  20  25  Koeleria macrantha cover (%)  10  Early  •  30  35  in southern interior  Figure  BC.  ) and Junegrass 3 2.03. Relationship between soil bulk density (Mg m (Koeleria macrantha) cover (%) at 20 rough fescue grassland treatment units  m  U  U  U) C U)  >..  0.6  >..  U) C U)  0)  100  (0 C II)  0)  Seral Stage  Figure 2.01. Relationship between soil bulk density (Mg m ) 3 and litter cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  50  1.2  0.0  0.2  0.4  Seral Stage  0  0)  0  0  0  5  0  o  •  15  (%)  20  2 y = 1E -05x 2 = 0.46** R n = 20  Exposed mineral soil  10  Early Mid Late  Seral Staae  0  +  +  25  0.Olx  0.78  30  5  Poa secunda cover (%)  10  15  20  Figure 2.02. Relationship between soil bulk density (Mg m ) 3 and exposed mineral soil (%) at 20 rough fescue grassland treatment units in southern interior BC.  0  0  .  47  Figure 2.04. Relationship between soil bulk density (Mg m ) and Sandberg’s 3 bluegrass (Poa secunda) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  0.0  0.2  0.4  0.6  0.8  1.0  1.2  0.0  0.2  0.4  0.6  0.8  000  0.8  0  1.0  1.0  0.6  0)  1.2  1.2  Ca  it’  a)  .2  LI)  11)0  0 00  0  a)  .2  0— 00 G)0  0  -  70 80  90 100  -  5  10  15  20  Poa secunda cover (%) Figure 2.07 Relationship between soil mechanical resistance (kPa) at 4.5 cm depth and Sandberg’s bluegrass (Poa secunda) mineral soil (%) at 20 rough fescue grasslandtreatment units in southern interior BC.  -  5  +  +  15  1008.5  Exposed mineral soil  10  65.03x  0  (%)  20  • o o  25  Early Mid Late  Seral  30  0  0  -  20  2 y= 0.1163x 2 = 0.28* R n = 20  .  -  •  +  40  1583  60 Festuca campestris cover  17.37x  0  (%)  I  I I  00  • 0 o  80  Early Mid Late  Seral  0  100  48  Figure 2.08 Relationship between soil mechanical resistance (kPa) at 4.5 cm depth and rough fescue (Festuca campestris) (%) at 20 rough fescue grasslandtreatment units in southern interior BC.  0  0  0  1000  500  a)  .2 c  1500  500  1000  11)0)  C)  0— 00  2000  2000  1500  2500  cu a  0  0  0  y = -1.0914x 2 = 0.34* R n =20  .  .  Figure 2.06 Relationship between soil mechanical resistance (kPa) at 4.5 cm depth and exposed mineral soil (%) at 20 rough fescue grassland treatment units in southern interior BC.  2500  Litter cover (%) Figure 2.05 Relationship between soil mechanical resistance (kPa) at 4.5 cm depth and litter cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  0  60  0  1000  1500  500  11)  .2 u  0— COO 11)0  500  1000  1500  2000  Cu  2000 a  2500  2500  0)  70  Litter cover (%)  80  90 100  0.00  0.05  0.10  0.15  0  0  0  •  5  Early Mid Late  Seral Staae  8  15  20  -  Exposed mineral soil (%)  10  0  y = -1E -04x 2 2 = 0.52** R n = 20 +  25  0.0007x  0.16  30  Figure 2.11. Relationship between fraction of total soil (kg kg ) in the 1-2 1 mm size class and exposed mineral soil (%) at 20 rough fescue grassland treatment units in southern interior BC.  (0(0 U-  o  8) 0(0 0)  ‘5  —(0 00  8) .C.  —  (N  E E  60  Figure 2.09. Relationship between fraction of total soil (kg kg ) in 1 the 1-2 mm size class and litter cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  (0(0 U-  .2  Ce  0 (08) —N . ‘5  -C —U)  —  E 0)  0) 0)  0.00  0.05  0.10  0.15  0.20  0  P  5  +  10  0.15  Poa secunda cover (%)  2 - 0.OOlx y = -0.0001x 2 = o.I* R n = 20  I  I  15  0  • o  Early Mid Late  Seral Staae  20  0) 0)  0.0  0.1  0.2  0.3  0.4  0.5  0  0  0  2  2 y = 0.0008x 2 = 0.37** R n = 20  •  +  +  6  0.31  0  I  8  Astragulus miser cover (%)  4  0.Olx  •  o  0  •  10  Early Mid Late  Seral Stage  12  49  Figure 112. Relationship between fraction of total soil (kg kg) in the 2-6 mm size class and timber milkvetch (Astragalus miser) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  U-  0  0)  —a)  i  —(0 00  CD (N 0)  E E  0.6  Figure 2.10. Relationship between fraction of total soil (kg kg) in the 1-2 mm size class and Sandberg’s bluegrass (Poa secunda) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  U-  0  0)  . .  —(0 00  (N  E E  0.25  E  I-  C)  C 0 .0  U  C) U)  I  0  C)  C 0 .0  0  0  E  0  8  0  2  4  6  8  10  0 00  5  .  15  Exposed mineral soil  10  I  I I  I  I I  I  (%)  20  • o o  Early Mid Late  25  Seral Stage  I  I I  I  I I I  30  60  Early Mid Late  •  80  Litter cover (%)  70  0  •  -  0  •  2 0.25x y = 0.002x 2 = 0.47** R n = 20  0  +  90  0  10.13  00  100  0  0  0  0  Figure 2.15. Relationship between total carbon (%) at 0-7.5 cm depth and litter cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  50  • o o  Seral Stage  Figure 2.13. Relationship between total carbon (%) at 0-7.5 cm depth and exposed ineral soil (%) at 20 rough fescue grassland treatment units in southern interior BC.  0  o o  y = -0.12x + 5.49 R = 0.37** n=20  C.)  E  0  2  4  6  0  .  5  0  Poa secunda cover  10  .  y = -0.14x + 5.26 2 = 0.28* R n20  (%)  15  0 0  •  Early Mid Late  Seral Stage  20  E C)  (U 0  F  C 0 .0 (U C)  U)  0  5  15  20  Koeleria macrantha cover (%)  10  25  30  50  Figure 2.16. Relationship between total carbon (%) at 7.5-15 cm depth and Junegrass (Koeleria macrant/ia) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  0  2  4  6  8  Figure 2.14. Relationship between total carbo (%) at 0-7.5 cm depth and s bluegrass(Poa secunda) cover (%) at 20 rough fescue 1 Sandberg grassland treatment units in southern interior BC.  U)  I-  (U 0  C 0  -o  14)  8  10  0.6  I—  0  C  2  C)  ci)  C  U)  U,  0  E  I-  C  C) 0  0  0  •  5 10  Poa secunda cover  y-0.Olx+0.43 2 = 0.28* R n=20  I I  I  (%)  o  O  •  15  Early Mid Late  Seral Stage  I  I I I I I  I  20  0  5  15  Exposed mineral soil  10  (%)  20  25  30  Figure 2.19. Relationship between total nitrogen (%) at 7.5-15 cm depth and exposed mineral soil (%) at 20 rough fescue grassland t reatment units in southern interior BC.  0.0  0.2  0.4  0.6  0.8  Figure 2.17. Relationship between total nitrogen (%) at 0-7.5 cm depth and Sandberg’s bluegrass(Poa secunda) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  0.0  0.2  0.4  0  0 It  :zE  0.8  I—  0  I-  C  0.0  0.2  0.8  0.0  0.2  0.4  0.6  Early Mid Late  70  0  (%)  80 Litter cover  • 0  90  - 0.03x 2 y = 0.0002x 2 = 0.50** R n = 20 +  1.21  100  0  0  0  0  5  15  20  Koeleria macrantha cover  10  (%)  25  30  35  Figure 2.18. Relationship between total nitrogen (%) at 0-7.5 cm depth and litter cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  60  • o o  Seral Stage  51  Figure 2.20. Relationship between total nitrogen (%) at 7.5-15 cm depth and Junegrass (Koeleria macranthai) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  C) 0  a)  C  0 U,  E  0.8  10 15 20 25  30  0  Koeleria macrantha cover (%)  0  20  40  0  .  0  0  0  2000  2 y -7E -07x 2 = 0.36* R n = 20  0 0  •. •  .  0  + +  46.28  6000  Litter biomass (kg ha ) 1  4000  0.008x  0  • o o  8000  Early Mid Late  Seral Stage  10000  ) 4 Figure 2.23. Relationship between soil polysaccharides (.tg m1 and litter biomass (kg had) at 20 rough fescue grassland treatment units in southern interior BC.  I—  a) 0  0  (U -C C) 0 (U () > 0 0.  a) V  60  80  20  40  60  80  0  0  1  2  +  3  59.58  4  Taraxacum officinale cover (%)  -  2 574x y = 0.23x 2 = 0.32* R n = 20  •  5  Early  Seral Stage  6  E  C a) a)  0) 11)  a) V  a) a)  E E  0  2  6  8  Astragulus miser cover (%)  4  10  12  52  Figure 2.24. Relationship between mean weight diamter (mm) and timber milkvetch (Astragalus miser) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  0.0  0.5  1.0  1.5  2.0  2.5  Figure 2.22. Relationship between soil polysaccharides (tg mr ) 1 and common dandelion (Taraxacum officinale) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  I.  a) 0  a)  0  0  (U 0  0 0.  0 0  20  a) >‘  >‘  CU (I)  Figure 2.21. Relationship between soil polysaccharides (.tg mH) and Junegrass (Koeleria macrantha) cover (%) at 20 rough fescue grassland treatment units in southern interior BC.  a)  35  8a)  0  40  a) -C  a)  0  a) V  0)  0)  a) V  a)  60  80  0 I—  cu  .0 Cu  0  >  a)  0  20  40  60  (%)  80  100  Festuca campestris cover  0  500  1000  1500  2000  2500  3000  0  0  o  00  20  .  40  0  60  Festuca campestris cover  y=11.38x+964.31 =0.41** 2 R n = 20  (%)  iI 0  0  •  80  Early Mid Late  Seral Stage  0  100  0  grassland treatment units in southern interior BC.  0  2  6  8  10  0  20  60  Festuca campestris cover  40  (%)  80  100  -J  .0  0  E  0) Co cci  0  2000  4000  6000  8000  10000  0  20  Early Mid Late  0  60  Festuca campestris cover (%)  40  0  0  +  0  80  2 + 101 .83x y = -0.72x 2 = 0.48** R n20 0  0  291.54  100  0  grassland treatment units in southern interior BC.  53  Figure 2.28. Relationship between litter biomass (kg had) and rough fescue (Festuca campesfris) cover (%) 20 rough fescue  °  • o  Seral Stage  (per m ) and rough fescue (Festuca campestris) cover (%) 20 rough fescue 2 grassland treatment units in southern interior BC.  Figure 2.26. Relationship between bunchgrass density (greater than 4 cm diameter)  Figure 2.27. Relationship between total aboveground living biomass (kg had) and rough fescue (Festuca campestris) cover (%) 20 rough fescue  2I:,)  D  >  .0  0  cu E  Co Co  0  Figure 2.25. Relationship between rough fescue (Festuca campestris) seedheads (per m ) and rough fescue cover (%) 20 rough fescue 2 grassland treatment units in southern interior BC.  LL  C)  c  I.  Co Co Cu  E  ci)  Co  >  A  c.  E  Cu  E  0.  0)  Cu C.) Cu C) D  4  8  10  12  0.  Co Co  0)  Co  0.  14  3. Effects of Cattle Grazing on Aggregate Stability on Rough Fescue Grasslands in the Southern Interior of British Columbia’ 3.1 Introduction Grasslands of the Pacific Northwest developed in a unique post-glacial environment of limited grazing. These grasslands have not experienced the prolonged, intense grazing by bison (Bison bison Linnaeus) that was a dominant feature of the Great Plains. In addition, populations of other wild ungulates (e.g., elk, mountain sheep (Ovis sp.), deer (Odocoileus sp.)) were also kept quite low by intense predation and aboriginal hunting pressure. This set of historical and ecological factors produced grassland species (such as bluebunch wheatgrass (Pseudoregoeneria spicata (Pursh) A. Love) and rough fescue(Festuca campestris Rydb.)) that have little resistance to continuous grazing, especially when compared to Euroasian and Great Plain grasses (Mack and Thompson 1982). When European settlers arrived in the southern interior of British Columbia (BC) in the mid 1 880s, they initiated cultivation, eliminated fire, introduced new grass species, and started confined grazing of cattle, often at intense grazing pressures (Campbell and Bawtree 1998). All those practices had profound impacts on grassland species composition and soil quality. The impacts on the former have been well documented (McLean and Wikeem 1985; Willms 1987; Willms et al. 1988; Milchunas and Lauenroth 1993; May et al. 2003; Short and Knight 2003), while changes of the latter have been less often investigated (van Ryswyk et al. 1966). Generally, modification of soil properties by grazing livestock is determined by many factors such as initial soil and vegetation status, climate, and type of grazing system including intensity, frequency, duration of grazing (Floate et al. 1981). High grazing intensity usually leads to reduced soil aggregate stability (Warren et al. 1986), increased soil compaction (Willatt and Pullar 1983; Chanasyk and Naeth 1995; Dormaar and Willms 1998), reduced water infiltration and enhanced soil erosion (Gifford and Hawkins 1979), and reduced soil fertility (Dormaar and Willms 1998). Soil organic matter is extremely important in grassland ecosystems since it improves soil structure (Thurow et al. 1986) and thereby enhances water infiltration (Smith et al. 1990) and reduction of soil erosion (Chevallier et al. 2001). Few studies have evaluated effects of grazing A version of this chapter will be submitted for publication. Lamagna, S.F., M. Krzic, R.F. Newman, B.M. Wallace, and G.E. Bradfield. Effects of cattle grazing on aggregate stability on rough fescue grasslands in the southern interior of British Columbia.  54  on soil organic matter and its relationship to aggregate stability (Martens and Frankenberger 1992; Dormaar and Wilims 1998; Abiven eta!. 2007). In general, findings are inconsistent depending on the type of organic matter compounds determined in a particular study. Some studies have found good correlations between carbohydrates content and soil aggregate stability (Haynes and Swift 1990; Anger eta!. 1993), while others have not (Carter and Kunelius 1993). The most likely reason for lack of agreement among the studies is that other soil organic matter compounds (Capriel et al. 1990) and fungal hyphae (Tisdale and Oades 1982) are involved in formation of stable aggregates. The objective of this study is to determine impacts of grazing on soil aggregate stability on the rough fescue grasslands of the southern interior of BC. The hypotheses are that (i) grazing reduces aggregate stability, soil C and N, polysaccharides, total aboveground living biomass, litter biomass, and litter cover, while increasing soil bulk density, exposed mineral soil, and mechanical resistance relative to rough fescue grasslands without grazing, (ii) aggregate stability is positively related to total C, C:N ratio, polysaccharides, litter biomass, and litter cover, and (iii) aggregate stability is negatively related to soil bulk density, soil mechanical resistance, and percent of exposed mineral soil.  3.2 Materials and Methods 3.2.1 Site Description The study was carried out on 18 treatment units consisting of ungrazed exclosures (between 0.5 and 1.0 ha in size) and adjacent grazed areas (Table 3.1). Treatment units were located at eight grassland sites within the Interior Douglas-fir biogeoclimatic zone in southern interior of BC. Treatment units were classified by seral stage (early, mid, late) of the ecosystem (Table 3.2). A treatment unit in an early seral stage had many forbs, invasive species, high percent of bare soil and low rough fescue density. A late seral stage included many bunchgrass species but most importantly, rough fescue. A mid seral stage usually included some rough fescue with bluebunch wheatgrass and other desirable grass species such as Idaho fescue (Festuca idahoensis Elmer). Eighteen grassland treatment units were classified by seral stage based on the percent cover of rough fescue using the following criteria (1) Early seral stage cover  <  10%, (2) Mid seral stage  =  =  rough fescue  rough fescue cover 11% 50%, and (3) Late seral stage = -  55  rough fescue cover> 50%. Five 30-rn long transects were randomly established both inside and outside of the exciosures and soil and vegetation sampling was done along these transects. All soil samples were collected in May 2007, while vegetation measurements and sampling were done in August 2007. Many rough fescue grasslands in BC have been grazed for over 100 years. The level of grazing varied among treatment units over time and is not reported here due to lack of verifiable data. Exclosures were established as early as 1931 and as late as 1994 (Table 3.1), while active grazing has continued on un-exclosed treatment units. We did not attempt to test the relationship between period of grazing protection (exclosure age) and rangeland health because of following confounding factors:  1. The ecological status of the sites, whether assessed by plant community or rangeland health, most likely differed widely at the time of exclosure construction. For example, there is some evidence that the Goose Lake exciosure was at a very early seral stage at the time of exclosure construction as a result of continuous heavy grazing. In contrast, some of the treatment units on the Lac du Bois grasslands were at mid to late seral stages at the time of exclosure construction. Therefore the starting point (e.g., plant community) for the expected recovery within the exclosures differed widely. 2. The level of grazing and grazing system used on the un-exclosed treatment units has differed among sites over time. Therefore, the grazing impact on plant community and rangeland health is expected to be different even for similar periods of grazing.  We proposed that the present seral stage of the plant community can be used as an integrating index of historical grazing use, including grazing use previous to exclosure construction. This is because cattle are known to have relatively consistent preferences for certain plant species. The preferred plant species will tend to decrease in the plant community with excessive grazing relative to non-preferred plant species following the “increaser/decreaser” model (Dyksterhuis 1949). The use of seral stage as an index to grazing relies on the following assumptions (i) succession in the rough fescue grasslands follows one pathway and (ii) multiple stable states do not occur.  56  3.2.2 Sampling and Analyses 3.2.2.1 Soil Sampling Intact soil samples for bulk density determination (Blake and Hartge 1986) were taken with a drop-hammer sampler (with a core of 7.5  x  7.5 cm) from 0-7.5 cm depth. One bulk  density sample was taken per transect within each treatment unit (i.e., five samples were collected per treatment unit). Soil samples for bulk density determination were dried at 105°C for 48 hours in a forced-air oven. Coarse fragments (diameter >2 mm) within the sample were screened out and weighed. Volume of mineral coarse fragments were determined from dry mass and assumed to have a particle density of 2.65 Mg m . Soil bulk density was calculated as the 3 mass of dry, coarse fragment free mineral soil per volume of field-moist soil, where volume was also calculated on a coarse fragment free basis. Soil mechanical resistance (Bradford 1986) was measured to 15 cm depth. Measurements were recorded at depth intervals of 1.5 cm, using a hand-pushed 13-mm diameter cone (3 0°) penetrometer with data logger (Agridry Rimik PTY Ltd., Toowoomba, QLD, Australia). Six soil profiles were recorded at random locations along each of the five transects totalling 30 measurements per treatment unit. The soil aggregate stability was determined using a variation of the wet sieving method (Nimmo and Perkins 2002). Five samples were taken at 0-7.5 cm depth at random locations along each transect and composited to make one per transect. Field moist samples were handsieved using a 6-mm sieve and collected on a 2-mm sieve. The pre-sieved 2-6 mm moist sample (of about 10 g) was placed on the top of a nest of sieves with openings of 2, 1, and 0.25 mm and wetted in a humidifier for about 60 minutes to minimize disruption caused by air trapping. This was done immediately before wet sieving. Wet sieving was performed for 10 minutes in a motor-driven mechanical device with a vertical stroke of 2.5 cm at a rate of 30 strokes per minute. The motion of the system had both a vertical stroke and an oscillating action through an angle of 30°. After the sieves were removed from the water the proportion of material retained on each sieve was oven dried at 105°C for 24 hours, weighed, and expressed as a percentage of the total soil. The results for aggregate stability were expressed as the mean weight diameter (MWD), which represents the sum of the mean diameter of each size fraction (Di) and the proportion of the sample weight occurring in the corresponding size fraction (Wi). The  57  summation was carried out over all four size-fractions, including the one that passed the 0.25mm sieve (MWD  =  ±  ). Corrections were made for non aggregate particles retained on 1 D  each sieve to avoid biased interpretations of water stable aggregates. Non-aggregate particles were subtracted from the mass of aggregates on each sieve by crushing all soil aggregates in a mortar and pestle then passing the material through the sieve and weighing the material that remained on the sieve. Samples were air-dried and ground to pass a 2.0-mm sieve. Total soil C (Nelson and Somers 1996) and total N (McGill and Figueirdeo 1993) were determined by elemental analysis using an automated analyzer (LECO CHN-600 Elemental Analyzer). Five soil samples per transect were taken at 0-7.5 cm depth and composited for soil C and N, soil polysaccharide and pH analysis. Soil polysaccharides were determined by the phenol-sulfuric acid method of Dubois eta!. (1956) as modified by Doutre eta!. (1978). Soil pH was determined on a 1:1 (v/v) soil to distilled water saturated paste and 1:1 (v/v) soil to 0.O1MCaC1 2 solution (Hendershot et al. 1993). 3.2.2.2 Plant Sampling Five portable cages (1 xl xl m) were installed per grazed treatment unit in May 2007 to exclude grazing by domestic and wild ungulates. Cages consisted of 1 m 2 meshed wire panels designed to permit light and precipitation in, but of a small enough mesh to prevent grazing. The portable cages were used to allow determination of annual growth on grazed treatment units. Living above-ground biomass was estimated from peak annual standing crop of grasses and forbs in August 1007. One-half m 2 areas were clipped to ground level and the plant litter was separated from current year’s growth of forbs and grasses. Five samples were randomly collected per treatment unit. Living plant material was sorted by species group, stored in paper bags and air-dried to minimize decomposition. Standing litter was retained and stored separately. Samples were oven-dried at 70°C to a constant weight, and weighed to the nearest 0.1 g. Fallen litter (detached above-ground dead plant material not incorporated into mineral soil) was collected from within the same 0.5m 2 areas used to determine living above-ground  58  biomass and combined with standing litter samples. Litter samples were processed in the same manner as living material. The density of key bunchgrasses, rough fescue, bluebunch wheatgrass, and Idaho fescue (Festuca idahoensis Elmer) were determined within five 1 m 2 plots per treatment unit. The number of flowering cuims of key bunchgrasses, rough fescue, bluebunch wheatgrass and Idaho fescue were counted within five 0.5 m 2 areas per treatment unit. The abundance of flowering cuims is related to reproductive capacity and plant recruitment potential and thus is an indication of the presence of functioning recovery mechanisms. 3.2.3 Statistical Analysis  Simple linear and second degree polynomial regression analyses were run. Treatment units were sampled five times to provide a better estimate of the mean. Univariate analysis of variance was conducted using seral stage as the single treatment factor to examine the effects of seral stage (i.e., grazing) on aggregate stability parameters (i.e., MWD and size distribution). The experimental design was an unbalanced incomplete block with eight blocks and six replicates of each treatment. The SAS procedure PROC MIXED was used for all analyses (SAS Institute 1989). Following a significant F-test, differences between treatment means were evaluated using Scheffe’s multiple comparison test. Results were considered significant at P < 0.10. Evidence from rough fescue grasslands of BC generally supports the single pathway, single equilibrium state. The only exception may be on the wettest sites where conversion from rough fescue to Kentucky bluegrass (Poa pratensis L.) has resulted in very slow recovery back to rough fescue. The complete loss of rough fescue is also expected to result in very slow recovery. Eighteen grassland treatment units were classified by seral stage based on the percent cover of rough fescue using the following criteria (1) Early seral stage = rough fescue cover < 10%, (2) Mid seral stage  =  rough fescue cover 11% 50%, and (3) Late seral stage -  =  rough fescue cover>  50%. These definitions of seral stages was used due to our sites being predominantly rough fescue. Rough fescue is a good indicator of the intensity of grazing on these particular sites.  59  3.3 Results and Discussion 3.3.1. Impacts of Grazing on Aggregate Stability Univariate analysis of variance was conducted using seral stage as the single treatment factor to examine the effects of seral stage (i.e., grazing) on aggregate stability parameters and several other soil and vegetation properties. Soil bulk density and mechanical resistance, litter cover and biomass, rough fescue cover and biomass, total aboveground living biomass, and exposed mineral soil were found to be sensitive to plant community seral stage, and therefore, by extension of our assumptions, to grazing (Table. 3.3). Difference always existed between early and late seral stages, but only occasionally between middle and late stages (e.g., rough fescue cover and biomass, total aboveground living biomass). The only statistically significant difference between early and mid seral stages was observed for litter cover and exposed mineral soil (Table 3.3). Lower litter cover was observed on the sites in early seral stage relative to those in mid seral stage, while the opposite trend was observed for exposed mineral soil. The benefits of using seral stage as an index of grazing history are that the grazing systems employed, grazing intensity, and grazing timing employed do not need to be quantified. It is assumed that seral stage is responsive to the cumulative impact of grazing. Furthermore, it is not necessary to control for the initial state of the plant community when discussing grazing impacts on range health. 3.3.2 Relationships Among Soil and Vegetation Properties with Aggregate Stability When aggregate stability parameters were correlated to either total C or N, the only strong relationships were observed between the proportion of the soil in the 1-2 mm size class and total soil C (R 2  =  0.62, Fig. 3.02) and total N (R 2 = 0.48, Appendix 3b). Since total C and N  data showed exactly the same trends with aggregate stability parameters, only total C data are presented in this chapter, while total N data are included in Appendix 3. Total soil C and N are indicators of the total soil organic matter, including a large proportion of stable, slowly decomposable organic compounds that contribute very little to aggregate formation. Consequently, it is not surprising that there were no particularly strong relationships between parameters of aggregate stability (with exception of the 1-2 mm aggregate size class) and total C or total N. The C :N ratio proved to be a bit of a better indicator of aggregate stability than total C (or N), since it explained 46% in variation of the 0.25-1 mm size class (Fig. 3.05) and 28% of  60  the 1-2 mm size class (Fig. 3.06). There was no significant correlation between the largest, 2-6 mm aggregates and the C:N ratio (Fig. 3.07). The same was true for the MWD (Fig. 3.08). Total soil C and N represent the total inventory of soil organic matter, which is considered to be a key attribute of soil quality and environmental quality (Gregorich et al. 1994; Carter 2002). Soil organic matter is involved in many soil chemical, biological, and physical properties such as susceptibility to compaction, friability, soil-water holding capacity, aggregation, potential water infiltration, nutrient supply, and susceptibility to erosion. Assessment of soil organic matter is a valuable step towards identifying the overall quality of a soil and whole ecosystem (Christiansen and Johnston 1997). Soil organic matter consists of a range of compounds from very stable to biologically active, including readily decomposable materials, litter and root biomass, and dead and living soil organisms. Total soil C and N provide a measurement of a soil’s total inventory of organic matter, while an array of properties (e.g., light fraction of the soil organic matter, mineralizable N, microbial biomass, carbohydrated, soil enzymes) reflect part of organic matter that is relatively easily decomposed (Gregorich et al. 1994). Among the indicators of aggregate stability used in this study, the proportion of aggregates in the smallest size fraction (0.25-1 mm) was best correlated (R 2  =  0.65) to soil  polysaccharides (Fig. 3.09), followed by the proportion of aggregates in the 2-6 mm size class 2 = 0.32, Fig. 3.11) and MWD (R (R 2  =  0.25, Fig. 3.12). There was no correlation between soil  polysaccharides and the 1-2 mm aggregate size class (Fig. 3.10). The strong correlation between the smallest aggregates and polysaccharides is in agreement with the current hierarchical model of aggregate formation and stabilization (Tisdall and Oades 1982). Large aggregates are comprised of an agglomeration of smaller aggregates, which are stabilized by various binding agents that range from plant roots and fungal hyphae (so called temporary binding agents) to polysaccharides (or transient binding agents). The importance of soil polysaccharides increases with reduction in aggregate size. A study done on Chernozemic soils under cultivation in southern Alberta by Dormaar (1987) showed that the smallest aggregate size class (< 100 .tm) measured was positively correlated with soil monosaccharides. The study also found that the soil monosaccharides in this size fraction were mostly of microbial origin rather than from plant debris. Another study done on a seeded pasture in Alberta by Dormaar et al. (2005) found that monosaccharides increased significantly (P < 0.00 1) with increasing aggregate size class.  61  Soil bulk density and mechanical resistance are parameters that are commonly used to describe compaction. When aggregate stability parameters were correlated to either soil mechanical resistance or bulk density, the better correlation was obtained with the soil mechanical resistance. Three out of four aggregate stability parameters had strong relationships with soil mechanical resistance (Figs. 3.17, 3.19, and 3.20), while only a proportion of the soil in the 1-2 mm size class was strongly correlated with soil bulk density (R 2  =  0.45, Fig. 3.14).  Strong, negative correlations were observed between MWD (and larger aggregatesO and soil mechanical resistance. In instances with a higher mechanical resistance, there is more compaction to the soil and will destroy or reduce aggregates. It affects the larger aggregates first since they are the most susceptible to compaction and destruction. Low soil bulk density and mechanical resistance are desirable because they imply no compaction and consequently, high water infiltration and good aeration creating a favourable environment for root growth. Generally, higher intensity grazing increases soil bulk density, decreases macroporosity, and destroys aggregates (Willatt and Pullar 1984; Warren et al. 1986; Chanasyk and Naeth 1995; Dormaar and Willms 1998; Vallentine 2001). The only strong, positive correlation between litter cover and any of the aggregate stability parameters was observed for the 1-2 mm aggregate size class (Fig. 3.22). This coincides with the strong (R 2 = 0.39), negative relationship between the 1-2 mm size class and exposed mineral soil (Appendix 4b). It appears that greater protection of the soil surface provided by the greater litter cover led to increase of the 1-2 mm size fraction. This particular size fraction also showed better correlation with soil properties such as soil bulk density, total C and N as well as C:N ratio, than other aggregate stability parameters. In addition, this was the only aggregate stability parameter that did not have a strong correlation with soil polysaccharides, a known binding agent of stable aggregates. Several other studies carried out on the grasslands of the southern interior of BC (Broersma et al. 2000; Krzic et al. 2000; Wallace et a!. 2008) showed that the 1-2 mm size fraction was consistently the least dominant fraction with less than 15% of the stable aggregates. This would support a hypothesis that 1-2 mm size aggregates are transient, mainly formed by destruction of larger aggregates (i.e., aggregates with diameter> 2 mm) or 2  aggregates grow to a critical size of 1  mm and then form larger aggregates (or compound  aggregates).  62  The aggregate size distribution determines (i) their susceptibility to movement (erosion) by wind and water and (ii) size of pores, which in turn affects the movement and distribution of water and air in the soil. When unstable aggregates are exposed to external stress (e.g., grazing, tillage) they easily break apart into small aggregates and primary mineral particles (i.e., clay, silt, sand) that then clog soil pores used for water drainage and gas exchange through the soil profile. The proportion of water stable aggregates in the 1—2 mm size class was best predicted by percent litter cover, Sandberg’s bluegrass cover, and percent of exposed mineral soil (Fig. 3.22, Fig. 3.30, and Appendix 4b), as compared to all other aggregate size fractions. Percent of litter cover had a positive relationship with the 1—2 mm aggregates (Fig. 3.22). Greater litter cover provided a better protection of aggregates at the soil surface and most likely provided better conditions for enhanced microbial activity. Both impacts would in turn enhance aggregate stability. The percent cover of Sandberg’s bluegrass was inversely related to the proportion of aggregates in the 1—2 mm size class (Fig. 3.30). Although the majority of treatment units had close to 0% Sandberg’s bluegrass cover, the remaining treatment units showed a consistent negative relationship with 1-2 mm aggregates. Sandberg’s bluegrass is an early seral species and establishes on sites with high grazing pressures. This relationship further emphasizes the need to investigate the usefulness of specific plant species, such as Sandberg’s bluegrass, as an indicator species for rangeland health. The proportion of aggregates in the 1—2 mm size class was inversely related to exposed mineral soil (Appendix 4b). This was in agreement with the relationship between litter cover and stable aggregates in this size class (Fig. 3.22). Sites with a high degree of exposed mineral soil generally have lower soil water content. Evaporation is encouraged from the lack of an insulating layer that plants and litter provide at the soil surface (Molinar et al. 2001). In grassland ecosystems that are generally facing a limited water supply, an increase in soil water content usually leads to increases in soil biological activity and plant vigour, which encourages the formation and stabilization of soil aggregates (Tisdall and Oades, 1982).  3.4. Conclusions Soil bulk density and mechanical resistance, litter cover and biomass, rough fescue cover and biomass, total aboveground living biomass, and exposed mineral soil were all found to be  63  affected by the plant community seral stage, and therefore, by extension of our assumptions, to grazing intensity. The 0.25-1 mm and 2-6 aggregate size classes and MWD did not correlate well with soil bulk density, soil mechanical resistance, total C and N, C:N ratio, litter cover, rough fescue cover, or exposed mineral soil. The 1-2 mm aggregate size class showed different relationships to soil/vegetation properties than the rest of the aggregate size class and MWD. This size fraction showed better correlation with soil bulk density, soil mechanical resistance, total C and N as well as C:N ratio, than other aggregate stability parameters. This size fraction was also the only aggregate stability parameter that did not have a strong correlation with soil polysaccharides, a known binding agent of stable aggregates. The 1-2 mm aggregates are probably formed from the destruction of larger aggregates and are, therefore, only temporary in those grassland sites. Further studies focusing on a variety of potential aggregate binding agents (hydrophobic aliphatic fraction, fungal hyphae) are needed. In addition, it would be useful to investigate if the 1-2 mm aggregate size class is also the least dominant class in other soil types besides Chernozems.  64  3.5 References Abiven, S., S. Menasseri, D.A. Angers, and P. Leterme. 2007. Dynamics of aggregate stability and biological binding agents during decomposition of organic materials. European Journal of Soil Science 58: 239-247. Angers, D.A. and G.R. Mehuys. 1993. Aggregate stability to water. p. 651-658. In M.R. Carter (ed.) Soil sampling and methods of analysis. Lewis Publisher, Boca Raton, FL. Angers, D.A., N. Bissonnette, A. Legere, and N. Samson. 1993. Microbial and biochemical changes induced by rotation and tillage in a soil under barley production. Canadian Journal of Soil Science 73: 39-50. Blake, G.R. and K.H. Hartge. 1986. Bulk density In A. Kiute (ed) Methods of soil analysis, Part 1. Physical and mineralogical methods-Agronomy monograph No. 9 (2fl’ edition), ASA SSSA. Madison, WI. Bradford, J.M. 1986. Penetrability. In A. Kiute (ed) Methods of soil analysis, Part 1. Physical and mineralogical methods-Agronomy monograph No. 9 (2Id edition), ASA-SSSA. Madison, WI. Broersma, K., M. Krzic, D.J. Thompson, and A.A. Bomke. 2000. Soil and vegetation of ungrazed crested wheatgrass and native rangelands. Canadian Journal of Soil Science 80: 411-417. Campbell, C.W. and A.H. Bawtree. 1998. Rangeland Handbook for BC. British Columbia Cattleman’s Association. Kamloops, BC. Capriel, P., T. Back, H. Borchert, and P. Harter. 1990. Relationships between soil aliphatic fraction extracted with supercritical hexane, soil microbial biomass and soil aggregate stability. Soil Science Society of America Journal 54: 4 15-420. Carter, M.R. and H.T. Kunelius. 1993. Effect of undersowing barley with annual ryegrasses or red-clover on soil structure in a barley-soybean rotation. Agriculture Ecosystems and Environment 43: 245-254. Carter, M.R. 2002. Soil quality for sustainable land management: organic matter and aggregation interactions that maintain soil functions. Agron. J. 94: 38-47. Chanasyk, D.S. and M.A. Naeth. 1995. Grazing impacts on bulk density and soil strength in the foothills fescue grasslands of Alberta, Canada. Canadian Jour. Soil Sci. 75: 551-557. Chevallier, T., E. Blanchart, and C. Girardin. 2001. The role of biological activity (roots, earthworms) in medium-term C dynamics in vertisol under a Digitaria decumbens (Gramineae) pasture. Applied Soil Ecology 16: 11-21.  65  Christensen, B.T. and A.E. Johnston. 1997. Soil organic matter and soil quality lessons learned from long-term experiments at Askov and Rothamsted. P. 399-430. In E.G. Gregorich and M.R. Carter (eds.) Soil quality for crop production and ecosystem health. Elsevier, Amsterdam. —  Dormaar, J.F. 1987. Quality and value of wind-movable aggregates in Chernozemic Ap horizons. Canadian Journal of Soil Science 67:601-607. Domaar, J.F and W.D. Willms. 1998. Effect of forty-four years of grazing on fescue grasslands. Journal of Range Management 51: 122-126. Dormaar, J.F., M.A. Naeth, W.D. Wilims, and D.S. Chanasyk. 2005. Effect of native prairie, crested wheatgrass (Agropyron cristatum (L.) Gaertn.) and Russian wildrye (Elymus junceus Fisch.) on soil chemical properties. Journal of Range Management 48: 258-263. Doutre, D.A., G.W. Hay, A. Hood, and G.W. VanLoon. 1978. Spectrophotometric methods to determine carbohydrates in soil. Soil Biol. Biochem. 10: 457-462. Dubois, M., K.A. Gilles, J.K. Hamilton, P.A. Rebers, and F. Smith. 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28: 350-356. Dyksterhuis, E.J. 1949. Condition and management of rangeland based on quantitative ecology. Journal of Range Management 2: 104-115. Floate, M.J.S., A. Rangeley, and G.R. Bolton. 1981. An investigation of problems of sward improvement on deep peat with special reference to potassium responses and interactions with lime and phosphorous. Grass and Forage Science 36: 8 1-90. Gayton, D.V. 2003. British Columbia grasslands monitoring vegetation change. F. Research Extenstion Partnership, Kamloops, B.C. Forrex Series 7. —  Gifford, G.F. and R.H. Hawkins. 1979. Deterministic hydrologic modeling of grazing system impacts on infiltration rates. Water Resources Bulletin 15: 924-934. Gregorich, E.G., M. R. Carter, D.A. Angers, C.M. Monreal, B.H. Ellert. 1994. Towards a minimum data set to assess soil organic matter quality in agricultural soils. Can J. Soil Sci. 74: 367-385. Haynes, R.J. and R.S. Swift. 1990. Stability of soil aggregates in relation to organic constituents and soil water content. Eur. J. Soil Science. 41: 73-83. Hendershot, W. H., H. Lalande and M. Duquette. 1993. Soil reaction and exchangeable acidity. In: M.R. Carter (ed) Soil Sampling and Methods of Analysis, Lewis Publishers, Boca Raton, FL pp. 141—145.  66  Krzic, M., K. Broersma, D.J. Thompson, and A.A. Bomke. 2000. Soil properties and species diversity of grazed crested wheatgrass and native rangelands. Journal of Range Management 53: 353-358. Mack, R.N. and J.N. Thomson. 1982. Evolution in steppe with few large, hoofed mammals. The American Naturalist 119: 757-773. Martens, D.A. and W.T. Frankenberg Jr. 1992. Decomposition of bacterial polymers in soil and their influence on soil structure. Biology of Fertile Soils 13: 65-73. May, K.W., W.D. Wiflms, and Z. Mengli. As assessment of variation in foothills rough fescue (Festuca campestris (Rydb.)) in southern Alberta. Canadian Journal of Plant Science 83: 541-550. McGill, W.B. and C.T. Figueiredo. 1993. Total nitrogen. p. 201-211 In M.R. Carter (ed.) Soil sampling and methods of analysis. Can. Soc. Soil Sci., Lewis Publ., Boca Raton, Fla. McLean, A. and S. Wikeem. 1985. Rough fescue response to season and intensity of defoliation. Journal of Range Management 38: 100-103. Milchunas, D.G. and W.K. Lauenroth. 1993. Quantitative effects of grazing on vegetation and soils over a global range of environments. Ecological Monographs 63: 328-366. Molinar, F., D. Galt, and J. Holechek. 2001. Managing for mulch. Rangelands 23:3-7. Nelson, D.W. and L.E. Sommers. 1996. Total carbon, organic carbon and organic matter. p. 961-1010 In D.L. Sparks et al. (eds.), Methods of soil analysis, Part 3, Chemical methods. No. 5 in SSSA Book Series, SSSA-ASA, Madison, Wisc. Nimmo, J.R. and K.S. Perkins. 2002. Aggregate stability and size distribution. p. 317-328. In A. J.H. Dane and G.C. Topp (ed.) Methods of soil analysis: Physical methods. Part 4. Number 5 in the Soil Science Society of America Book Series. Madison, Wise. SAS Institute Inc. 1989. SAS/STAT user’s guide. Version 6, 4th ed., Vol.2. SAS Inst. Inc., Cary, N.C. Smith, H.J.C., G.J. Levy, and I. Shainberg. 1990. Water-droplet energy and soil amendments effect on infiltration and erosion. Soil Science Society of America Journal 54: 1084-1087. —  Short, J.J. and J.E. Knight. 2003. Fall grazing affect big game forage on rough fescue grasslands. Journal of Range Management 56: 213-217. Thurow, T.L., W.H. Blackburn, and C.A. Taylor, Jr. 1986. Hydrologica characteristics of vegetation types as affected by livestock grazing systems, Edwards Plateau, Texas. Journal of Range Management 39: 505-509.  67  Tisdale, J.M., and J.M. Oades. 1982. Organic matter and water-stable aggregates in soils. J. of Soil Sci. 33:141—163. Vallentine, J.F. 2001. Grazing management. Academic Press. San Francisco, CA. Van Ryswyk, A.L., A. McLean, and L.S. Marchand. 1966. The climate, native vegetation, and soils of some grasslands at different elevations in British Columbia. Canadian Journal of Plant Science 46: 35-50. Wallace, B.M., M. Krzic, T.A. Forge, K. Broersma, and R.F. Newman. 2008. Biosolids increase soil aggregation and protection of soil C five years after application on a crested wheatgrass pasture. Journal of Environmental Quality (in press). Warren,S.D., M.B. Nevill, and W.H. Blackburn. 1986. Soil response to trampling under intensive rotation grazing. Soil Science Society of America Journal 50: 1336-1341. Wilims, W.D. 1987. Response of rough fescue (Festuca scabrella) to light, water, temperature, and litter removal, under controlled conditions. Canadian Journal of Botany 66: 429-434. Wilims, W.D., J.F. Dormaar, and G.B. Schaaije. 1988. Stability of grazed patches on rough fescue grasslands. Journal of Range Management 41: 503-508. Willatt, S.T. and D.M. Pullar. 1984. Changes in soil physical properties under grazed pastures. Australian Journal of Soil Research 22: 343-348.  68  50° 3’ 49”  120°25’ 46”  50° 6’ 12”  120°25’ 37”  Latitude  Longitude  438  3.4  Estimated Annual Precip. (mm)  Estimated Annual Temp. (°C)  TW/TH/GW  TE/TH  28  3.4  -  MS  14  4.8  379  1000  15  120°26’ 1”  50° 48’ 41”  1981  ..  GW  18  3.4  438  1175  4.5  120°51’ 56”  50° 35’ 56”  1993 & 1963  Tunkwa Lake  1920, 1931, & 1994  TM  27  3.4  438  1035  5  120°40’ 26”  AY  19  4.8  379  920  7  120°26’ 56”  50° 47’ 12”  1978  Drum Lake  50° 35’ 36”  Agriculture Canada Weather Station -  MQ  9  4.8  379  900  15  120°22’ 49”  50° 47’ 35”  1981  Deep Lake Fertilizer 1  27”  TH  27  3.4  438  1306  2  120°25’ 31”  5004  1970’s  Microwave Repeater  69  ‘TW: Medium textured morainal deposits, slightly to very stony; TH: Medium and moderately fine textured morainal deposits, slightly to moderately stony; GW: Coarse textured fluvioglacial deposits with moderately coarse or medium textured cappings, moderately to very stony; MS: Medium textured morainal deposits, slightly to moderately stony; AY: Gravelly, coarse and moderately coarse textured ice-contact (ablation moraine) deposits, moderately to very stony; MQ: Medium and moderately fine textured morainal deposits, slightly to moderately stony; TM: Medium and moderately fine textured morainal deposits, slightly to moderately stony.  Parent Material  (%)  Avg. Coarse Fragments  33  1240  1160  Elevation(m) 438  5  Slope  5  (%)  1968  1931  Date Established  Summit  Goose Lake  Descnption  Lac du Bois Fertilizer 1  Site  Table 3.1. Location, site characteristics, and year of exciosure establishment of eight sites in the southern interior of British Columbia  Table 3.2. Soil textural classes and seral stage of 18 study treatment units located in the southern interior of British Columbia. Description Soil % Seral % % Site Treatment Unit Texture Sand Silt Clay Stage Goose Lake  Grazed Ungrazed  Loam Loam  41 34  17 20  42 46  Early Late  Summit  Grazed Ungrazed  Loam Loam  38 37  20 21  42 42  Early Late  Grazed Ungrazed  Sandy loam Loam  74 48  6 14  20 38  Mid Mid  Grazed Ungrazed Old Ungrazed New  Loam Loam Loam  45 39 43  18 24 22  37 37 35  Early Late Mid  Grazed Ungrazed -1 Ungrazed -2  Loam Loam Loam  37 38 45  16 21 16  47 41 39  Early Mid Mid  Agriculture Canada Weather Station  Grazed Ungrazed  Loam Sandy loam  48 52  13 18  39 30  Mid Mid  Deep Lake Fertilizer 1  Grazed Ungrazed  Sandy loam Loam  52 50  18 16  30 34  Mid Late  Microwave Repeater  Grazed Ungrazed  Sandy loam Loam  61 37  15 27  24 36  Early Late  Lac du Bois Fertilizer 1 -  Tunkwa Lake  -  -  Drum Lake  -  70  0.169 0.273 0.556 0.239 0.192 0.9 12 0.271 0.039 0.100  0.16 (0.01)a 0.15 (0.02)a 0.37 (0.04)a 5.3 (0.44)a 0.43 (0.04)a 12.6 (0.33)a 60.8 (3.7)a 0.76 (0.03)b 983.7 (164.6)b  0.15 (0.01)a 0.11 (0.02)a 0.38 (0.04)a 4.0 (0.50)a 0.32 (0.04)a 12.7 (0.34)a 51.2 (4.4)a 0.88 (0.04)ab 1255.4 (194.8)ab  0.13 (0.01)a 0.16 (0.02)a 0.33 (0.04)a 4.2 (0.49)a 0.32 (0.04)a 12.6 (0.34)a 51.1 (4.0)a 0.98 (0.04)a 1744.9 (177.8)a  Fraction of soil in 1-2 mm aggregate size class (kg kg’) Fraction of soil in 0.25-1 mm aggregate size class (kg kg’)  Fraction of soil in< 0.25 mm aggregate size class (kg kg’)  (%) at 0-7.5 cm depth Total soil N (%) at 0-7.5 cm depth  C:N ratio  Soil polysaccharides (rig mL’) ) 3 Soil bulk density (Mg m  Soil mechanical resistance (kPa)  0.040 0.066 0.008 0.046 0.010  95.5 (3.4)b 4718.0 (778.5)b 74.1 (5.8)b 2025.5 (196.2)b 1682.7 (153.2)b  93.9 (3.9)b 1102.9 (874.4)ab 23.4 (6.8)a 991.6 (221.2)a 342.5 (177.5)a  74.9 (3.7)a 598.9 (867.8)a 1.8 (6.3)a 875.4 (216.6)a 83.8 (164.9)a  Litter biomass (kg ha-’) Festuca campestris cover (%)  Total aboveground living biomass (kg haj  Festuca campestris biomass (kg ha’)  0.054 1.54 (2.5)b 5.3 (2.7)b 13.1 (2.7)a Exposed mineral soil (%) Values in the same row followed by a same letter are not significantly different (P <0.10) according to Scheffe’s test. Values in brackets are the standard error of the mean (n = 6).  Litter cover  (%)  Total soil C  0.709  0.33 (0.04)a  0.38 (0.04)a  0.34 (0.04)a  ) 1 Fraction of soil in 2-6 mm aggregate size class (kg kg-  71  Table 3.3 Soil aggregate stability parameters, total C and N, C:N, soil polysaccharides, soil bulk density, mechanical resistance, exposed mineral soil, and selected vegetation parameters in response to seral stage on 18 treatment units of rough fescue grasslands in the southern interior of British Columbia. Seral Stage Property P Late Mid Early 0.766 1.7 (0.15)a 1.9 (0.17)a 1.7 (0.16)at MWD (mm)  Cl) 0  0.6  0.00  0.05  0.10  0.15  0.20  LI  0)  .-  0) Ow 0) 0)  00) CON  —o  .  Cl) 0 CC3  a)  CD  2  0  •.  0  Total Carbon  4  0  00  Oo  (%)  6  .0 0  •  8  Early Mid Late  Seral Staae  0  +  -  0  • o  1  Early Mid Late  2  0  •  0 0  0  •  3  4  0  0  •—  0.19x -0.03  Seral Stacie  2 y = -0.02x 2 = 0.09 R n=18  5  6  Figure 3.01. Relationship between total carbon (%) and the 0.25-1 mm aggregate size class (kg kg ) at 18 rough fescue 1 grassland treatment units in southern interior BC.  0  -  2 0.04x + 0.22 y = 0.006x 2 = 0.06 R n = 18 0  7  10  Total Carbon (%) Figure 3.03. Relationship between total carbon (%) and the 2-6 mm aggregate size class (kg kg’) at 18 rough fescue grassland treatment units in southern interior BC.  0.0  0.1  0.2  0.3  0.4  E E 0.5  p  CON  ow  —C)  —  .C  a)—  E E  0.25  U  a) E  U)  I  E E  I  a) 0  G)  .2  0) 0 00) —N  — 0 U)  E  0  2  0  o0  6  0 0  0  .  Total Carbon (%)  4  0  2 + 0.llx -0.12 y = -0.01x 2 = 0.62*** R n = 18  0  0  •  8  Early Mid Late  Seral Stage  10  0.0  0.5  1.0  1.5  2.0  2.5  0  +  2  3  0.84x -0.03  4  0  5  0  0  •  6  Early Mid Late  Seral Stage  7  72  Total Carbon (%) Figure 3.04. Relationship between total carbon (%) and mean weight diameter (mm) at 18 rough fescue grassland treatment units in southern interior BC.  1  2 y -0.09x 2 = 0.12 R n = 18  •  o  aggregate size class (kg kg ) at 18 rough fescue grassland treatment 1 units in southern interior BC.  Figure 3.02. Relationship between total carbon (%) and the 1-2 mm  0.00  0.05  0.10  0.15  0.20  0.25  0.5  0.6  ia..  0  ‘  CO  °aj  •o  0.0  0.1  0.2  0.3  11.0  0  -  Early Mid Late  13.0  13.5  9  e  2.5  0.00  0.05  11.0  0  11.5  o  0  •  Early Mid Late  Seral Stage  0  12.0  5.50  12.5  0  •  0  13.0  •  •0  13.5  00  14.0  0  0  11.5  0  12.0  0 0  •  C:N  12.5  Early Mid Late  Seral Stage  13.0  -  13.5  2 2.23x y = 0.09x 2 = 0.31* R n=20  0 +  14.0  14.36  0  6  I  I  I I I  I  11.5  0  °  •  12.0  Early Mid Late  Seral Stage  0  I  I  I I I  I  C:N  12.5  -  0  0  13.0  +  13.5  2 9.60x y = 0.38x 2 = 0.35* R n20  •  •  61.90  14.0  0  73  Figure 3.08. Relationship between C:Nand mean weight diameter (mm) at 18 rough fescue grassland treatment units in southern interior BC.  0.0 11.0  0.5  1.0  2.0  0  Figure 3.06. Relationship between C:Nand the 1-2 mm  aggregate size class (kg kg ) at 18 rough fescue grassland 1 treatment units in southern interior BC.  12.5  IL  0.10  015  +  C:N  12.0  0.33  o5  =  .9  0.20  -  = 0.04 0.87x R = 2 0.29* n18  C:N  +  0  9  0.25  Figure 3.05. Relationship between C:N and the 0.25-1 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  11.5  2 0.06x y = 0.003x =0.55*** 2 R n = 18  0 0  •  Seral Stage  Figure 3.07. Relationship between C:Nand the 2-6 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  11.0  0.00  0.05  0.10  0.15  0.20  0.4  c.’J  cç  IL  O)  oo  =.3  E E  0.25  E E  .  .  Ca  0)  0)  LL  0) ca Ca  c  -  o  —m  Co.!  —o  .c ci —ø cm  Co  U-  c  O)N  O)  —  -  ci c cm —o  o.)  E E  60  Polysaccharides (pg  50  .  ) 1 mr  o o  •  70  Early Mid Late  Seral Stage  0  80  .  0)  ._-.  Ca 0)  U-  .2 ca  Ca)  0  ‘.-  C)  —Co  a) -C  E  E  0.00  0.05  0.10  0.15  0.20  0.25  30  +  0.11  0  0  40  60  ) 1 Polysaccharides (pg mr  50  •  o o  •  70  Early Mid Late  Seral Stage  •-z-o•  y = 0.0006x 2 = 0.04 R n = 18  80  60  0  70  0.10  80  i  E  0.0  i.s  p2.0  30  .  0  • 0  40  Early Mid Late  Seral Stage  .  0  50  0  =  n  18  +  0  -0.0006x 2 0.250 =  60  =  2 R  y  0  70  0.05x  +  0.95  80  74  Figure 3.12. Relationship between soil polysaccharides (jig mH) and mean weight diameter (mm) at 18 rough fescue grassland treatment units in southern interior BC.  50  0  +  ) 4 Figure 3.11. Relationship between soil polysaccharides (jig m1 and the 2-6 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  40  Mid  0 0  Late  Early  •  Seral Stage  00  0.Olx  Polysaccharides (pg mr ) 1  30  .  0  +  Polysacchandes (pg mr ) 1  0.0  0.1  0.2  0.3  0.4  0.5  0.0002x y 2 2 = 0.32* R n18 =  2.5  0  -  0.02x 0.43  0.6  40  +  Figure 3.10. Relationship between soil polysaccharides (jig m1 ) 1 and the 1-2 mm aggregate size class (kg kg’) at 18 rough fescue grassland treatment units in southern interior BC.  30  0  2 y = -0.0001x 2 = 0.65*** R n = 18  Figure 3.09. Relationship between total polysaccharides (jig ml) and the 0.25-1 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  0.00  0.05  0.10  0.15  0.20  0.25  0)  t:  0)0)  .2  0)  2a  U) —N  —U)  .  (N  (9)  0) 0) c0 (U U-  —0)  —o 00)  0)..— .C (0 — (0  to ‘0) (N  E E  1.0 1.2  U-  0)0)  .2  0) CO)  E E  0.00  005  0.10  0.4  • o •  0.6  Early Mid Late  Seral Stage  0.33  0.8  1.0  1.2  0.4  +  0.6  3.43x  Early Mid Late  2 y -1 .85x 2 = 0.13 R n = 18  =  • o •  -  0.8  0  ) 3 Bulk density (Mg m  1.20  0  00  cP .  1.0  1.2  ) and the 3 Figure 3.15. Relationship between bulk density (Mg m 2-6 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  0.0  0.1  0.2  0.3  0.4  0.5  Seral Stage  C (U ‘1)  a)  0)  (U 0  E  0) 0)  0.4  •  0  •  0.6  Early Mid Late  0.8  0  Bulk density (Mg rn ) 3  0  1.0  •  oi>Z  13.73x -4.36  Seral Stage  +  1.2  75  ) and mean 3 Figure 3.16. Relationship between bulk density (Mg m weight diameter (mm) at 18 rough fescue grassland treatment units in southern interior BC.  0.0  0.5  1.0  1.5  2.0  2 y = -7.54x 2 = 0.12 R n18  2.5  0.8  0  .  0.15  0.20  -  0.6  0.6  Early Mid Late  0  0  2 + 1 .22x y = -0.77x 2 = 0.45** R n = 18  ) and the 3 Figure 3.14. Relationship between bulk density (Mg m 1-2 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  • o •  Seral Stage  18  0.78  Figure 3.13. Relationship between bulk density (Mg m ) and the 3 0.25-1 mm aggregate size class (kg kg’) at 18 rough fescue grassland treatment units in southern interior BC.  =  +  Bulk density (Mg rn ) 3  0.4  n  -  2 1 .47x y = 0.83x 2 = 0.22° R  0.25  Bulk density (Mg rn) 3  0.00  0.05  0.10  0.15  0.20  0.25  0.20  C)  0)  —w  . .  aj  =  (U °  (N  (9  Li..  00)  o o  600  .  0  800  0 0  •  1000  I  Early Mid Late  Seral Stage  0  o°”  0  1200  0  1400  I  0  0  I  1600  1800  -  +  2000  0  1 E-07x 2 0.0004x  Mechanical resistance (kPa)  0  =  R = 0.32* 2 n = 18  y  2200  0.38  2400  600  800  +  1000  y = -1E-07x 2 2 = 0.42** R n = 18  0 0  0  0  1800  • 0 0  2000  I  Early Mid Late  Seral Stage  •  Mechanical resistance (kPa)  I  1600  I  1200 1400  0.0006x -0.12  0  2200  2400  Figure 3.19. Relationship between mechanical resistance at 7.5 cm (kPa) and the 2-6 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  0.0  01  0.2  0.3  0.4  0.5  0.6  Figure 3.17. Relationship between mechanical resistance at 7.5 cm (kPa) and the 0.25-1 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  0.00  0.05  0.10  0.15  00)  L))  E  E  0.25  Li..  .0)  0  j  øa)  —(  (;l’)  E E  a)  C CU  0)  E  ..  20  0.0  0.5  1.0  1.5  .  =  800  0 0  0  +  +  1400  0 0  •  I  1600  Early Mid Late  Seral Stage  0  0.09  II  I  I  II I  1800  Mechanical resistance (kPa)  1200  I  I I  I  1 E-4x  1000  2 4EO8x  R = 0.07 2 n = 18  y  2000  0  2200  2400  800  • 0 0  1000  Early Mid Late  1400  1600  1800  2 y = -7E-07x 2 = 0.38** R n = 18  +  Mechanical resistance (kPa)  1200  Seral Stage  •  0  0  2000  2200  0.002x - 0.05  0  2400  76  Figure 3.20. Relationship between mechanical resistance at 7.5 cm (kPa) and mean weight diameter (mm) at 18 rough fescue grassland treatment units in southern interior BC.  600  0 0  0  0  Figure 3.18. Relationship between mechanical resistance at 7.5 cm (kPa) and the 1-2 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  600  2.5  0.00  0.05  0.10  0.15  0.20  0.25  -  (U 0  .  .  C) 0 C) CU  -  —CU 00  :.  -  (N C)  co  E  E  (U U-  :  ° C.-. 0 C)  (U0  = 0  —  -C  a).  ‘4) C) C•4 CC)  E E  0.0  0.1  0.2  0.3  0.4  0.5  0.6  80  0  0  90  0  0  100  E C)  0  U-  C)  C)  a) OCU  (U  —CU 00  a)  C’.)  E  E  E  0.00  0.05  0.10  0.15  0.20  50  • o 0  .  60  Early Mid Late  Seral Stage  70  .  -  .0  80  90  y = 1 E-5x 2 0.000lx 2 = 0.39** R n = 18  •  +  0.07  000  100  60  Early Mid Late  I -  0  80  90  2 0.Olx y = 8E-5x =0.02 2 R n = 18  Litter cover (%)  70  .  .  +  0.82  100  C  00  0  Figure 3.23. Relationship between litter cover (%) and the 2-6 mm aggregate size class (kg kg ) at 18 rough 1 fescue grassland treatment units in southern interior BC.  0  0  •  Seral Stage  0  0  00  a)  C (U  C)  (U 0  E  a)  0.0  0.5  1.0  1.5  2.0  50  80  Litter cover (%)  70  0  0  •  90  Early Mid Late  Seral Stage  0  100  0  Figure 3.24. Relationship between litter cover (%) and mean weight diameter (mm) at 18 rough fescue grassland treatment units in southern interior BC.  60  y0.002x+ 1.54 2 = 0.009 R n = 18  •  0  00  Litter cover (%)  70  0.42  Figure 3.22. Relationship between litter cover (%) and the 1-2 mm aggregate size class (kg kg ) at 18 rough 1 fescue grassland treatment units in southern interior BC.  60  Early Mid Late  +  Litter cover (%)  • o o  Seral Stage  -  2 0.007x y = 4E-5x 2 = 0.03 R n=18  0.25  Figure 3.21. Relationship between litter cover (%) and the 0.25-1 mm aggregate size class (kg kg ) at 18 rough 1 fescue grassland treatment units in southern interior BC.  50  50  0.00  0.05  0.10  0.15  0.20  0.25  77  a)  0)  0) ( 0) LL  .2  0 C 0)  ‘—  o  U) —a)  —o  .c  cD. CJ 0)  E  U-  0 0) 0)  00)  caca  °a,  0  5  10  15 20  25 30  35  0  5  y = -0.0O05x 2 = 0.20 R n = 18  0  +  +  15  0.31  20  • o o  30  Early Mid Late  Seral Stage  25  Koeleria mcrantha cover (%)  10  0.Olx  .  35  Figure 3.27. Relationship between Junegrass (Koeleria macrantha) cover (%) and the 2-6 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  0.0  0.1  0.2  0.3  0.4  0.5  0.6  Figure 3.25. Relationship between Junegrass (Koeleria macrantha) cover (%) and the 0.25-1 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  Koeleria macrantha cover (%)  U-  .2  0.00  o.o5  0.10  0.15  0.20  0 5  0.16  15  +  20  25  Koelena macrantha cover (%)  10  -  2 0.004x y = 7E-05x 2 = 0.21° R n = 18  0 0  •  30  Early Mid  Seral Stage  35  a)  C  0) a)  E  a) a)  E E  0  5  y = -0.002x 2 2 = 0.16 R n = 18  +  +  15  1.61  20  25  Koeleria macrantha cover (%)  10  0.05x  • o o  30  Early Mid Late  Seral Stage  35  78  Figure 3.28. Relationship between Junegrass (Koeleria macrantha) cover (%) and mean weight diameter (mm) at 18 rough fescue grassland treatment units in southern interior BC.  0.0  0.5  1.0  1.5  2.5  Figure 3.26. Relationship between Junegrass (Koeleria macrantha) cover (%) and the 1-2 mm aggregate size class (kg kg) at 18 rough fescue grassland treatment units in southern interior BC.  0 C 0)  •U) I!  = 0  .  a).  c’J  E  00)  Lf) .;)  E E  0.25  0.0  0.1  0.2  0.3  0.4  0.5  0.6  0  0  0  5  .  10  Poa secunda cover (%)  0  2 - 0.0004x y = 0.0001x =0.01 0 2 R n18 +  Mid Early  0 0  15  Early  •  Seral Stage  0.35  20  Figure 3.31. Relationship between Sandberg’s bluegrass (Poa secunda) cover (%) and the 2-6 mm aggregate size class (kg kg 1) at 18 rough fescue grassland treatment units in southern interior BC.  .  . )  00 ci)  -; a)N  C. .— 0 .  .  CO c  t:  20  0.00 0  •  5  10  I  15  0 0  •  Early Mid Early  SeraiStage  I I  I  20  C a) ci)  a)  0)  a)  E E a) a) E  0.0  0.5  1.0  1.5  2.0  2.5  5  +  1.76  10  Poa secunda cover (%)  2 - 0.009x y = 0.0007x 2 = 0.004 R n = 18  •  0  0  15  0  0  •  Early Mid Early  Seral Stage  •  20  79  Figure 3.32. Relationship between Sandberg’s bluegrass (Poa secunda) cover (%) and mean weight diameter (mm) at 18 rough fescue grassland treatment units in southern interior BC.  0  0  0  Figure 3.30. Relationship between Sandberg’s bluegrass (Poa secunda) ) at 18 4 cover (%) and the 1-2 mm aggregate size class (kg kg rough fescue grassland treatment units in southern interior BC.  15  0.05  0.10  0.15  0.20  I I  I  Poa secunda cover (%)  10  LL  a)a)  .2  a) 0 0) Ca)  ..-  . —N  ;  C a)  a)c -Co  0)  y=-O.0001x 2 O.OOlx+O.15 =O.34* 2 R n18  Figure 3.29. Relationship between Sandberg’s bluegrass (Poa secunda) cover (%) and the 0.25-1 mm aggregate size class (kg kg ) at 18 4 rough fescue grassland treatment units in southern interior BC.  0.00  0.05  0.10  0.15  0.20  0)  E_. C  0.25  Poa secunda cover (%)  [—  a) 0)  oi C 0) .2  00 a)  —a) a)  0-  Lf) 0) c’J .‘ 00) a). .C c  E E  0.25  4. General Conclusions 4.1 Research Conclusions British Columbia’s grasslands are among the most diverse in North America because of the wide variations in climate, soils, topography, and historical use (Tisdale 1947). This exceptional diversity makes classification and management of the grasslands very challenging. Out of the 94 million hectares of land area in British Columbia, the ranching industry uses 10 million hectares (8.5 million hectares is Crown land). Grasslands make up about 10% of British Columbia’s rangelands (Wikeem et al. 1993), while the remaining 90% is under either open forest or occur in clearcuts (with and without forage seeding). British Columbia’s grasslands are unique ecosystems, consisting of species and habitats that contrast sharply with the North American prairies. Grasslands of British Columbia contain more than 30% of the provincial wildlife species of concern, and support more threatened or endangered species than any other ecosystem. This is in agreement with a general recognition that grasslands are critical for maintenance of global biodiversity (Gayton 2003). Yet, between 76 and 99% of British Columbia’s native grasslands have been altered either due to agricultural production, inappropriate livestock grazing practices, urban expansion, invasive weeds, abusive recreational activities, forest encroachment, or climate change (Campbell and Bawtree 1998). Consequently, sustainability of grasslands is important for both biodiversity maintenance and the ranching industry. British Columbia’s grassland regions have also been recognized internationally for recreation and tourism opportunities. To our knowledge, no studies have evaluated the linkages among indicators of rangeland health assessments and quantitative measures of plant and soil properties in British Columbia and only a limited number of studies have focused on quantification of these relationships anywhere in the world. A recently published manuscript by Miller (2008) evaluated the status of three ecosystem attributes (soil/site stability, hydrologic fhnction, and biotic integrity) using the qualitative assessment protocol “Interpreting Indicators of Rangeland Health” across a broad spatial extent of 760,000 ha in Grand Staircase  —  Escalante National Monument, Utah. That  study found that sites with the greatest production potential were the most degraded and that net effects of past management practices have not been ecologically beneficial, indicating that ongoing management needs to be site specific accounting for site sensitivity to degradation.  80  Since rough fescue grasslands on British Columbia are very sensitive to grazing, they are well suited for studies that will focus on the establishment of relationships between rangeland health assessment indicators and soil and vegetation properties. Cattle grazing has affected all nine sites that were studied, but some more severely than others. Each site was categorized into early, mid, or late seral stage depending on the amount of rough fescue present on the site. Whether or not seral stage is a good indicator of rangeland health is still debated. Seral stage has many problems associated with it including (i) the climax of an ecosystem is not always the most desirable condition, (ii) pristine conditions may not be the actual climax for a site, (iii)it does not allow for exotic species, and (iv) it is not well suited for woodland and forested rangeland (Smith 1978; Miller 2008). However, the benefits of using seral stage as an index of grazing history are that the grazing systems employed, grazing intensity, and grazing timing employed do not need to be quantified. It is assumed that seral stage is responsive to the cumulative impact of grazing. Furthermore, it is not necessary to control for the initial state of the plant community when discussing grazing impacts on range health (Gayton 2003). In this study, total C and N had very similar correlations with exposed mineral soil, percent litter cover, percent rough fescue cover, percent Sandberg’s bluegrass cover, and percent Junegrass cover. Total C and N had a positive relationship with percent litter cover and percent rough fescue cover while they had a negative relationship with exposed mineral soil, Sandberg’s bluegrass cover, and Junegrass cover. Soil polysaccharides related the closest with the smallest aggregate size fraction (0.25-1 mm) which was in agreement with the current hierarchical model of aggregate formation of Tisdale and Oades (1982) that polysaccharides bind together smaller aggregates, while roots bind larger aggregates. Twenty-five potential indicators were evaluated for their ability to predict rough fescue grassland health as defined by the National Research Council’s three principal criteria. The top five rangeland health indicators based on strength of relationship (R ) with vegetation and soil 2 properties (known to be affected by grazing) were: percent exposed mineral soil, percent litter cover, percent Sandberg’s bluegrass cover, percent Junegrass cover, and percent rough fescue cover. Percent rough fescue was particularly important in this study since it was the only indicator of the presence of functioning recovery mechanisms. There was also some redundancy when the top five indicators were further analyzed. Sandberg’s bluegrass had a similar trend 81  with exposed mineral soil showing a positive relationship. Exposed mineral soil also had a strong, negative relationship with litter cover showing that either exposed mineral soil or litter cover should be used. Percent rough fescue cover was found to be an essential indicator of the presence of functioning recovery mechanisms. There were no substitute indicators found that can be used to assess this particular health category. Percent exposed mineral soil was a sensitive indicator of the degree of soil stability and watershed function, as well as an indicator of the integrity of nutrient cycles and energy flows in rough fescue grasslands. Percent Junegrass cover was not as sensitive an indicator as percent exposed mineral soil, but has general overall strength with many health measures. These observations show that there is a consistency among easily-assessed rangeland health indicators and quantitative vegetation and soil properties that are linked with properties that are more difficult to measure. With these findings, a set of indicators can be established throughout the rangeland community; farmers, ranchers, and government officials can use these indicators to assess the health of the rangeland in the southern interior of British Columbia. These findings can also be used for rough fescue rangeland found in other places around the world.  4.2 Evaluation of Study Methods and Recommendations for Future Research This study was carried out in the interior Douglas-fir biogeoclimatic zone in British Columbia during the 2007 growing season (vascular plant cover and microbiotic crust measurements were done during the 2006 growing season). Studies on development and application of criteria and indicators for forests and grasslands are often lacking, or have been done on a limited number of sites with relatively narrow ranges of climate and soil type (Working Group on Criteria and Indicators for Conservation and Sustainable Management of Temperate and Boreal Forests 1999; Page-Dumroese et al. 2000). To our knowledge, no studies have evaluated the linkages among indicators of rangeland health assessments and quantitative measures of plant and soil properties in British Columbia and only a limited number of studies have focused on quantification of these relationships anywhere in the world (Miller 2008). This study showed that there are distinct linkages between easily-assessed rangeland health indicators and soil/vegetation properties.  82  This study also showed that aggregate stability was a key property that correlated well with many other soil and vegetation properties (e.g., soil polysaccharides, total C, total N, bulk density, mechanical resistance, Sandberg’s bluegrass cover, and exposed mineral soil). This is important since aggregate stability has been used as an indicator of soil quality in various ecosystems, including rangelands. The purpose of a soil quality assessment is to determine the ability of soil to perform a desired set of functions. Within the rough fescue grasslands in the southern interior of British Columbia, the critical soil functions are to decrease the susceptibility to soil compaction, the susceptibility of species invasion, and provide an area for desirable species to grow. Soil compaction is an extremely important property to monitor on rough fescue grasslands since rough fescue cannot do very well with a high amount of soil compaction. Bulk density is commonly measured by collecting intact soil cores. It was relatively easy to measure bulk density in the field even though there were rocks present, although it was much more time consuming than soil mechanical resistance. An advantage to using bulk density as an indicator of soil compaction was that it is relatively cheap if there are soil cores available and there are ovens and scales available for use. Although bulk density has been shown to be less sensitive to soil compaction than mechanical resistance (Chanasyk and Naeth 1995; Krzic et a!. 1999), there were similar significant differences found between seral stages in relation to bulk density and soil mechanical resistance. Although soil mechanical resistance can provide more data and is quicker to use, there are some drawbacks. The penetrometer’s steel rod must be inserted into the soil at a constant velocity. With this particular study, we had one person doing these measurements so that there was not a lot of variability with the force onto the soil surface and down through the soil profile. There seemed to be a consistency with the data (and along with the bulk density data) that it can be assumed that one person should always do the penetrometer measurements to keep it consistent. Total soil C and N are often listed as key indicators of soil quality, since they represent the total inventory of soil organic matter, which is considered to be a key attribute of soil quality and environmental quality (Gregorich et al. 1994; Carter 2002). Soil organic matter is involved in many soil chemical, biological, and physical properties such as susceptibility to compaction, friability, soil-water holding capacity, aggregation, water infiltration, nutrient supply, and  83  susceptibility to erosion. Assessment of soil organic matter is a valuable step towards identifying the overall quality of a soil and whole ecosystem (Christensen and Johnston 1997). Soil organic matter consists of a range of compounds from very stable to biologically active, including readily decomposable materials, litter and root biomass, and dead and living soil organisms. As mentioned above, soil total C and N provide a measurement of a soil’s total inventory of organic matter, while an array of properties (e.g., light fraction of the soil organic matter, mineralizable N, microbial biomass, carbohydrates, soil enzymes) reflect part of organic matter that is relatively easily decomposed (Gregorich et al. 1994). Carbohydrates in general, and polysaccharides in particular, contribute to soil and grassland health through their role in the stabilization of soil structure and providing a readily available source of energy for soil organisms. The collection of total C and N samples are relatively straightforward and can be carried out by myriad different people and on many levels of experience. The cost for total C and N analysis was about $10 per sample. Measuring soil polysaccharides was time consuming; however, it was a valuable property to measure to enhance the aggregate stability data. Further research should be conducted on the 1-2 mm aggregate size class in determining its sensitivity to other soil and vegetation properties. The 1-2 mm aggregate size class related differently than the rest of the aggregate size class and mean weight diameter for most of the correlations. This size fraction showed better correlation with soil properties such as soil bulk density, soil mechanical resistance, total C and N as well as C:N ratio, than other aggregate stability parameters. This size fraction was also the only aggregate stability parameter that did not have a strong correlation with soil polysaccharides, a known binding agent of stable aggregates. The 1-2 mm aggregate size class is probably formed from the destruction of larger aggregates and are, therefore, only temporary in the soil. This temporary stage seems to be most sensitive to soil and vegetation properties. All vegetation properties measured were easy to measure and relatively cheap. All that was needed were ovens and scales which were donated to us by Agriculture and Agri-Food Canada in Kamloops, BC. Collecting the vegetation data was much quicker than collecting the soil properties data. Litter cover, rough fescue cover, Junegrass cover, Sandberg’s bluegrass cover, and exposed mineral soil were the top five properties that were best correlated with all the other soil and vegetation properties. It would be important to note the redundancy of certain properties; Sandberg’s bluegrass demonstrated similar trends with exposed mineral soil and  84  exposed mineral soil demonstrated similar trends with litter cover. Hence, in future health assessments as well as studies, indicators that do not contribute unique (new) information should not be done. Timber milkvetch, in particular, needs further research done on rough fescue grasslands. Generally, greater percentage of timber milkvetch cover indicates lower ecological status (e.g., lower seral stage) of a grassland. Since this study showed that timber milkvetch increases with increasing aggregate stability, it goes against what is usually seen (Majak et al. 1996). Timber milkvetch is a leguminous plant species that fixes N from the atmosphere. Sites with high cover of this species may be able to support a larger soil microbial population (Biederbeck et al. 2005), which would in turn lead to greater aggregate stability. It is possible that this enhanced N availability and microbial activity due to a greater presence of timber millkvetch obscured negative grazing impacts on aggregate stability that could be expected on sites with greater cover of this plant species. It has been observed that timber milkvetch initially decreases with grazing, and as grazing pressure intensifies cover of this species increases (Majak et al. 1996). This relationship requires further study because it was not an expected outcome. Multiple regressions may be useful in some cases to better understand the relationship between certain parameters. In particular, the relationship between rough fescue cover and soil mechanical resistance should be looked at in more detail to take bluebunch wheatgrass, Kentucky bluegrass, or common dandelion into account. The information gathered from this study can help farmers, ranchers, government employees, and the general public in understanding the rough fescue grassland ecosystem. The easily assessed indicators allow for a wide range of people and levels of experience to investigate the land and determine its overall health or quality. This enhances communication among interested groups of society that have otherwise been mis-communicating. There needs to be an understanding both from the government perspective and the local rancher perspective that both need to compromise and understand what each one is trying to do. Further research should be done on the effectiveness of using seral stage as an indicator for grazing intensity since there are drawbacks to it. Overall, there has to be an increase in communication between researchers, the government, and local farmers and ranchers in order to keep rough fescue grasslands in good rangeland health.  85  4.3 References Biederbeck, V.0., R.P. Zentner, and C.A. Campbell. 2005. Soil microbial populations and activities as influenced by legume green fallow in a semiarid climate. Soil Biol. Biochem. 37:1775—1784. Campbell, C.W. and A.H. Bawtree. 1998. Rangeland Handbook for BC. British Columbia Cattleman’s Association. Kamloops, BC. Carter, M.R. 2002. Soil quality for sustainable land management: organic matter and aggregation interactions that maintain soil functions. Agron. J. 94: 38-47. Chanasyk, D.S. and M.A. Naeth. 1995. Grazing impacts on soil bulk density and soil strength in the foothills fescue grasslands of Alberta, Canada. Canadian 3. of Soil Sci. 75: 55 1-557. Christensen, B.T. and A.E. Johnston. 1997. Soil organic matter and soil quality lessons learned from long-term experiments at Askov and Rothamsted. P. 399-430. In E.G. Gregorich and M.R. Carter (eds.) Soil quality for crop production and ecosystem health. Elsevier, Amsterdam. —  Gayton, D.V. 2003. British Columbia grasslands monitoring vegetation change. F. Research Extenstion Partnership, Kamloops, B.C. Forrex Series 7. —  Gregorich, E.G., M.R. Carter, D.A. Angers, C.M. Monreal, B.H. Ellert. 1994. Towards a minimum data set to assess soil organic matter quality in agricultural soils. Can J. Soil Sci. 74: 367-385. Krzic, M., R.F. Newman, K. Broersma, and A.A. Bomke. 1999. Soil compaction of forest plantations in interior of British Columbia. Journal of Range Management 52: 67 1-677. Majak, W., L. Stroesser, J.W. Hall, D.A. Quinton, and H.E. Douwes. 1996. Seasonal grazing of Columbia milkvetch by cattle on rangelands in British Columbia. Journal of Range Management 49: 223-227. Miller, M.E. 2008. Broadscale assessment of rangeland health, Grand Staircase-Escalante National Monument, U.S.A. Rangeland Ecology and Management 61:249-262. Page-Dumroese, D., Jurgensen M., Elliot W., Rice T., Nesser J., Collins T. and Meurisse R. 2000. Soil quality standards and guidelines for forest sustainability in northwestern North America. For. Ecol. Manage. 138:445-462. Smith, E.L. 1978. A critical evaluation of range condition concept. p. 266-267. In: D.N. Hyder (ed.). Proceedings First International Rangeland Congress. Society for Range Management. Denver, CO.  86  Tisdale, E.W. 1947. The grasslands of the southern interior of British Columbia. Ecology 28: 346-382. Tisdale, J.M., and J.M. Oades. 1982. Organic matter and water-stable aggregates in soils. J. of Soil Sci. 33:141—163. Wikeem, B.M., A. McLean, A. Bawtree, and D. Quinton. 1993. An overview of the forage resource and beef production on crown land in British Columbia. Canadian Journal of Animal Science 73: 779-794. Working Group on Criteria and Indicators for Conservation and Sustainable Management of Temperate and Boreal Forests. 1999. Montreal Process.  87  Appendices Appendix 1. Negative correlations among potential indicator variables (independent) and attributes deemed to be related to good site health (deendant variable). Dependant variable  Potential indicator  Correlation coefficient  Fraction of soil in 1 2 mm aggregate class (kg kg’) Bulk density (Mg rn) 3 Total carbon (7.5 15 cm depth) (%) Total carbon (0 7.5 cm depth) (%)  Exposed mineral soil (%) Litter cover (%) Koeleria macrantha cover (%) Koeleria macrantha cover (%) Koeleria macrantha cover (%) Exposed mineral soil (%) Litter cover (%) Exposed mineral soil (%) Exposed mineral soil (%) Koeleria macrantha cover (%) Exposed mineral soil (%) Exposed mineral soil (%) Litter cover (%) Pseudoroegneria spicata small plant density (per m ) 2 Crustose lichen cover (%) Koeleria macrantha cover (%)  -0.70 -0.67 -0.63  -  -  -  Total nitrogen (7.5 15 cm depth) (%) Total nitrogen (7.5 15 cm depth) (%) Mechanical resistance (kPa) at 4.5 cm soil depth Total nitrogen (0 7.5 cm depth) (%) Total carbon (0 7.5 cm depth) (%) Total nitrogen (0 7.5 cm depth) (%) Total carbon (7.5 15 cm depth) (%) Soil water content (%) Pseudoroegneria spicata 2 small plant density (per m ) 2 Fraction of soil in 1 2 mm aggregate class (kg kg-’) -  -  -  -  -  -  -  Fraction of soil in 1 2 mm aggregate class (kg kg-’) Fraction of soil in 0.25 1 mm aggregate class (kg kg-’) -  -  Festuca campestris ‘big plant seed heads (per m ) 2 Festuca campestris cover (%) Festuca campestris biomass (kg ha-’) Fraction of soil in 1 2 mm aggregate class (kg kg-’) Bulk density (Mg m ) 3 Mechanical resistance (kPa) at 4.5 cm soil depth Fraction of soil in 1 2 mm aggregate class (kg kg-’) -  -  Bulk density (Mg m ) 3 Festuca campestris biomass (kg ha-’) Total carbon (7.5 15 cm depth) (%) -  Mechanical resistance (kPa) at 4.5 cm soil depth Festuca campestris cover (%)  Silt (%) Other grass biomass (kg ha-’) Forbs biomass (kg ha-’) Poa secunda cover (%) Rosa acicularis cover (%) Litter biomass (kg ha-’) Pseudoroegneria spicata small plant seed heads (per m ) 2 Litter biomass (kg ha-’) Koeleria macrantha cover (%) Pseudoroegneria spicata small plant seed heads (per m ) 2 Festuca campestris biomass (kg has) Forbs biomass (kg ha-’)  -0.63 -0.62 -0.61 -0.61 -0.60 -0.60 -0.60 -0.60 -0.60 -0.60 -0.59 -0.59 -0.58 -0.58 -0.58 -0.57 -0.57 -0.56 -0.56 -0.55 -0.54 -0.54 -0.54 -0.53 -0.53  88  Dependant variable  Potential indicator  Correlation coefficient  Total nitrogen (0 7.5 cm depth) (%) Total carbon (0 7.5 cm depth) (%) Soil water content (%) Total carbon (0 7.5 cm depth) (%)  Poa secunda cover (%) Poa secunda cover (%) Poa secunda cover (%) Pseudoroegneria spicata small plant seed heads (per m ) 2 Other grass biomass (kg ha’)  -0.53 -0.52 -0.52 -0.52  -  -  -  Festuca campestris biomass (kg ha-’) Mechanical resistance (kPa) at 4.5 cm soil depth Total nitrogen (7.5 15 cm depth) (%) -  Total nitrogen (7.5 15 cm depth) (%) Fraction of soil in < 0.25 mm aggregate class (kg kg’) -  ‘big = greater than 4 cm diameter; 2 small  =  Festuca campestris cover (%) Pseudoroegneria spicata small plant seed heads (per m ) 2 Poa secunda cover (%) Astragalus miser cover (%)  -0.52 -0.51 -0.51 -0.51 -0.51  less than 4 cm diameter  89  Appendix 2. Positive correlations among potential indicator variables (independent) and attributes deemed to be related to good site health (dependant variable). Dependant variable  Potential indicator  Correlation coefficient  Total carbon (7.5 15 cm depth) (%) Total nitrogen (7.5 15 cm depth) (%) Total carbon (0 7.5 cm depth) (%) Total nitrogen (0 7.5 cm depth) (%)  Soil water content (%) Soil water content (%) Soil water content (%)  0.91 0.91 0.90 0.89  -  -  -  -  Exposed mineral soil (%) Litter biomass.(kg ha-’) Total nitrogen (7.5 15 cm depth) (%) Total nitrogen (0 7.5 cm depth) (%) Total nitrogen (7.5 15 cm depth) (%) Total nitrogen (0 7.5 cm depth) (%) Mechanical resistance (kPa) at 4.5 cm soil depth Pseudoroegneria spicata big plant seed heads (per m ) 2 ) 2 Pseudoroegneria spicata 2 small plant seed heads (per m Forbs biomass (kg ha-’) Exposed mineral soil (%) Bulk density (Mg rn) 3 Soil water content (%) Pseudoroegneria spicata small plant seed heads (per m ) 2 -  -  -  -  Total nitrogen (7.5 15 cm depth) (%) Total carbon (7.5 15 cm depth) (%) ) 2 Pseudoroegneria spicata small plant density (per m Total carbon (0 7.5 cm depth) (%) Bulk density (Mg m ) 3 Litter cover (%) Total carbon (7.5 15 cm depth) (%) Total carbon (7.5 15 cm depth) (%) ) 2 Pseudoroegneria spicata small plant density (per m ) 2 Pseudoroegneria spicata bigplant seed heads (per m Pseudoroegneria spicata small plant seed heads (per m ) 2 Sand (%) Total nitrogen (7.5 15 cm depth) (%) ) 2 Pseudoroegneria spicata small plant density (per m Total carbon (0 7.5 cm depth) (%) Mechanical resistance (kPa) at 4.5 cm soil depth -  -  -  -  -  -  -  Soil water content (%) Poa secunda cover (%) Festuca campestris biomass (kg ha-’) Allium cernuum cover (%) Rosa acicularis cover (%) Rosa acicularis cover (%) Allium cemuum cover (%) Poa secunda cover (%) Allium cernuum cover (%) Poa secunda cover (%) Koeleria macrantha cover (%) Crustose lichen cover (%) Exposed mineral soil (%) Poa pratensis cover (%) Pseudoroegneria spicata small plant density (per m ) 2 Litter cover (%) Litter cover (%) Exposed mineral soil (%) Rosa acicularis cover (%) Poa secunda cover (%) Soil water content (%) Rosa acicularis cover (%) Allium cemuum cover (%) Crustose lichen cover (%) Other grasse biomass (kg ha’) Forbs biomass (kg ha’) Astragalus miser cover (%) Poa pratensis cover (%) Poa secunda cover (%) Allium cemuum cover (%) Festuca idahoensis small plant seed heads (per m ) 2  0.87 0.81 0.74 0.73 0.72 0.72 0.69 0.69 0.68 0.68 0.68 0.68 0.67 0.66 0.66 0.65 0.65 0.65 0.65 0.64 0.64 0.64 0.64 0.64 0.63 0.63 0.62 0.62 0.62 0.62  90  Dependant variable  Potential indicator  Correlation coefficient  Total nitrogen (0 7.5 cm depth) (%) Total carbon (0 7.5 cm depth) (%) Fraction of soil in 1 2 mm aggregate class (kg kg-’) Total carbon (7.5 15 cm depth) (%) Fraction of soil in I 2 mm aggregate class (kg kg-’) Total nitrogen (0 7.5 cm depth) (%) Soil aggregate mean weight diameter (mm) Fraction of soil in 2 6 mm aggregate class (kg kg-’) Other grass biomass (kg ha-’) Mechanical resistance (kPa) at 4.5 cm soil depth  Litter cover (%) Litter cover (%)  0.62 0.61  Litter cover (%) Poa pratensis cover (%) Total carbon (7.5 15 cm depth) Poa pratensis cover (%) Astragalus miser cover (%) Astragalus miser cover (%) Poa pratensis cover (%) Exposed mineral soil (%)  0.61 0.61 0.60 0.60 0.60 0.60 0.60 0.59  Total carbon (0 7.5 cm depth) (%) Bulk density (Mg rn) 3 Mechanical resistance (kPa) at 4.5 cm soil depth Festuca idahoensis cover (%)-small plant density (per m ) 2 Pseudoroegneria spicata small plant seed heads (per m ) 2 Pseudoroegneria spicata small plant seed heads (per m ) 2 Mechanical resistance (kPa) at 4.5 cm soil depth Fraction of soil in 1 2 mm aggregate class (kg kg-’) Festuca campestris big plant seed heads (per m ) 2 Other grass biomass (kg ha-’) Festuca campestris small plant density (per m ) 2  Poa pratensis cover (%) Crustose lichen cover (%) Soil pH at 0 7.5 cm depth (CaCl ) 2 Poa secunda cover (%) Festuca idahoensis cover (%) Exposed mineral soil (%) Soil pH at 0 7.5 cm depth Total carbon (0 7.5 cm depth) (%) Sand (%) Allium cemuum cover (%) Soil pH at 0 7.5 cm depth (CaCl ) 2 Koeleria macrantha cover (%) Bryophytes cover (%) Soil pH at 0 7.5 cm depth Soil pH at 0 7.5 cm depth (CaCl ) 2 SoilpHato-7.5cmdepth Forbs biomass (kg ha-’) Crustose lichen cover (%) Soil water content (%) Festuca campestris cover (%)  -  -  -  -  -  -  -  -  -  Bulk density (Mg m) 3 Fraction of soil in 2 6 mm aggregate class (kg kg-’) Festuca campestris small plant density (per m ) 2 Festuca idahoensis small plant seed heads (per m ) 2 Festuca idahoensis small plant seed heads (per m ) 2 Mechanical resistance (kPa) at 4.5 cm soil depth Pseudoroegneria spicata small plant seed heads (per m ) 2 Fraction of soil in 1 2 mm aggregate class (kg kg-’) Total aboveground living biomass (kg ha’) -  -  ‘big  =  greater than 4 cm diameter; 2 small  =  -  -  -  -  -  -  -  (%)  0.59 0.58 0.58 0.57 0.56 0.56 0.56 0.55 0.54 0.54 0.53 0.52 0.52 0.51 0.51 0.51 0.50 0.50 0.50 0.41  less than 4 cm diameter  91  CU  0.6  0.00  -  ‘5  CU  0) CU  .2 0)  .  Cl)  .Co  1)  0.0  0.1  0.2  0.3  0.0  0.0  o.os  o.io  0.15  0.20  0.25  E 05 E  c)  g  00)  00)  ...  i •5  U  , ,;0)  ‘.—.  E  E  a)  II I  I  0 0  •  0.2  Mid Late  Early  Seral Stage  0.2  Late  Mid  0  0  Early  •  •  Seral Stage  I  II  II  I cP  0.4  0  0.4  0  Total nitrogen  0  • 0  Total nitrogen  0  0  0.6  -  (%)  0.6  +  0.21  +  0.20  ___—ô  0.29x  + 0.85x 2 y = -1.09x 2 = 0.15 R = 20  (%)  0  2 y = 0.27x =0.09 2 R n = 20  0.8  0.8  .  •  ‘5  0)  CU  E  a)  E E  d)  .20)  00)  Co  0  CoO) —N  —Co  E E  b)  0.0  0.5  1.0  1.5  2.0  2.5  0.00  0.05  0.10  0.15  020  0.25  0.0  0.0  I  I  I  I  I I  II  I  I  •  0.2  Early Mid Late  0 0  •  0.2  Early Mid Late  Seral Stage  0  0  •  Seral Stage  I  I  I  I  I I  I I I  I  0.4  0  00  0.4  0  Total nitrogen  0  Total nitrogen  0  0  00  (%)  =  0.6  -5.06x 4.17x + 2 0.18 n = 20 =  2 R  y  (%)  0.6  +  -  0.97  y = -O.62x 2 + 0.65x 0.006 2 = 0.48** R n=20  92  0.8  0.8  Appendix 3. Relationship between total nitrogen (%) and the aggregate size class (kg kg ) a) 0.25-1 mm, b) 1-2 mm, c) 2-6 mm, and 1 d) mean weight diameter (mm)  IL  (I)  LI-  0)  gz5  0.5  E E  0.0  0.1  0.2  0.3  0.4  0.6  0.00  0.05  0.10  0.15  0.20  0.25  c)  0  _o  _O)  E E  a)  0  0  0  0  0  o  5  5  o  0 •  0  0  o  0•  -  +  0.14  15 20  15  Exposed mineral soil  10  •  20  (%)  II  I  I  o  o  •  25  .  0.35  o  • 0  25  Early Mid Late  Seral Staoe  .  +  Early Mid Late  Seral Stage  y = -0.0009x 2 = 0.006 R n=20  Exposed mineral soil (%)  10  2 0.004x y = 0.0002x 2 = 0.14 R n=20  30  30  ( LL  ci)  a  0)  a  E  E i5  d)  D)  o  =e  E E  b)  1.0  1.2  1.4  1.6  1.8  2.0  2.2  2.4  2.6  0.00  0.05  0.10  0.15  0.20  0.25  0  0  0  00  •  o  5  5  0  15  10  15  Exposed mineral soil (%)  0  Exposed mineral soil (%)  10  25  Early Mid Late  20  25  y=-0.007x+ 1.77 R=0.02 n = 20  0 0  •  Seral Stage  20  Appendix 4. Relationship between exposed mineral soil (%) and the aggregate size class (kg kgj a) 0.25-1 mm, b) 1-2 mm, c) 2-6 mm, and d) mean weight diameter (mm).  93  30  30  .  00) ! (a  0)  ,  —.  o•Z5  = (a  I  -  2 0.0005x y = -8E-06x =0.05 2 R n = 20 +  0.14  0.5  0.6  0.00  0.05  0.10  0.0  0.1  0.2  03  0  9 c  0  I  I  I II  I  20  o  0 0  •  20  I  40  Q  I  60  I  I I  II 0 0  •  II  I  40  = +  0  80  0.000lx  cover (%)  60  -6E-06x 2 = 0.03 fl20  y  Festuca campestris  Early Mid Late  I  I  0  80  +  Early Mid Late  I  I  I  0.36  Seral Stage  Festuca campesfns cover (%)  Seral Stage  0  0  0  o  100  100  0.15 9-—o  0.20  025  04  E  E  c)  00)  00)  •.E w  U) 0) (N 00)  E E  a)  2.5  0.00  0.05  o.o  0.15  •  a,  C (a  0)  0.0  Q5  1.0  1.5  2.0  d)  0) 00) ‘  00  E  0.25  E c- 0.20  b)  0  0  0  0  0  20  0  2 y = -2E-05x =0.23 2 R n=20  0  +  +  0.13  60  I  I  0  Fesfuca campesfns cover (%)  40  0.002x  o  80  0 0  •  Early Mid Late  Seral Stage  0  0  I  100  I I  I  I  I I  I  20  0 0  •  I  I  I  I  I  40  60  2 y = -4E-05x 2 = 0.02 R n20  Festuca campestris cover (%)  Early Mid Late  I Seral Stage I  0  +  0  80  0.002x  +  100  1.75  L.——-—__0  •  •  0  94  Appendix 5. Relationship between rough fescue (Festuca campestris) cover (%) and the aggregate size class (kg kg-’) a) 0.25-1 mm, b) 1-2 mm, c) 2-6 mm, and d) mean weight diameter (mm).  Appendix 6. Detailed description of nine study sites within the Interior Douglas-fir biogeoclimatic zone in the southern interior of British Columbia. Goose Lake Goose Lake is located in the Nicola Valley, about 17 km east of Merritt, BC (Appendix 6). The exciosure on this site was established in 1931 and is considered to be in its late seral stage with rough fescue as the climax species. The site is in excellent rangeland health condition inside the exclosure. The grazed area outside the exclosure is in very poor condition (one of the poorest among all the sites that were sampled). There were lots of rocks on the surface with big patches of exposed mineral soil as well as several plant species (yarrow, aster, Kentucky bluegrass, Junegrass, needle-and-thread). This site had the most distinct differences between the grazed and ungrazed areas. Drum Lake Drum Lake is located in the Nico la Valley, about 7 1cm southeast of Merritt, BC  (Appendix 6). There are two exciosures on this site. There were lots of disturbance due to voles inside both of the exciosures as well as on the grazed area (however much less outside the exciosure). These exclosures were of medium quality with a few small rough fescue plants as well as some bluebunch wheatgrass. There was lots of Junegrass as well as Sandberg’s bluegrass. There were several spotted knapweed plants in both the grazed and ungrazed areas. Hamilton Summit’ Hamilton Summitt is located in the Nicola Valley, about 18 km southeast of Merritt, BC and just south of the Microwave Repeater site (Appendix 6). The exelosure on this site was established in 1968. The ungrazed exciosure is in excellent is in excellent rangeland health condition. The grazed are is in poor condition with lots of exposed mineral soil and lots of rocks. There were lots of Junegrass and Poa secunda seedheads outside the exclosures. Microwave Repeater Microwave Repeater is located in the Nicola Valley, about 18 km east of Merritt, BC and just south of the Goose Lake site (Appendix 6). The exclosure at this site is in excellent quality with heaps of rough fescue and a huge litter layer. The soil at this site is Orthic Black Chernozem. This exelosure does not belong to the Range Reference Area and it was established as the exclosure for the receptor at the site. The grazed area is of medium to poor quality with some small rough fescue plants and other non-bunchgrasses (Sandberg’s bluegrass, Junegrass, Kentucky bluegrass, needle-and-thread). 95  Tunkwa Lake  Tunkwa Lake is located about half way between Merrit, BC and Kamloops, BC and is close to the Tunkwa Lake resort (Appendix 8). This site is at a lower elevation from the other forested grazed sites. There are two exclosures at this site (new and old). The old exciosure has more rough fescue than the new exclosure. On the new exciosure there was an extensive coverage of bromegrass and crested wheatgrass, which were seeded to improve rangeland condition. There was some gopher disturbance on both exclosures, however more on the new exclosure. The grazed area was in medium health condition similar to the new exciosure. Lac du Bois Fertilizer  —  1  Lac du Bois Fertilizer  —  1 is located in the Lac du Bois Provinicial Park, north of  Kamloops, BC (Appendix 7). It has a relatively shallow soil since it is located at the top of a drumlin. The exciosure is in medium to very good rangeland health conditions. The grazed area is in medium to poor rangeland health condition. Lac du Bois Fertilizer  —  6  This site is also located at the Lac du Bois Provinicial Park and together with the Lac du Bois Fertilizer  -  1 is part of the Agriculture Canada grazing intensity trial (Appendix 7). The  exciosure is in medium rangeland health condition with a patch of roses in the middle of the exciosure. The grazed area has good vegetation coverage but very little grasses. It was heavily disturbed in the past, hence lots of non-grass species (wild lupine, lemonweed, wild onion) moved in. The grazed area is considered to be in poor rangeland health condition. Deep Lake Fertilizer  —  1  The grazed and ungrazed area at Deep Lake Fertilizer  —  1 were very similar (Appendix  7). The exclosure is located on a slope and it looked a little greener from a distance. There was good rough fescue coverage but also lots of balsam root. There was relatively shallow soil since the exclosure is on a slope. The grazed area was also in fairly good condition. There was some delphinium (a small, blue, poisonous plant) on the grazed area as well. Agriculture Canada Weather Station  Agriculture Canada Weather Station is located in Lac du Bois Provincial Park, north of Kamloops, BC (Appendix 7). It is very close to Lac du Bois Lake. It is located on a drumlin (formed by deposited debris at the edges of glaciers and is usually granularly textured); hence it has a very shallow soil (the thinnest among all sites) and coarse texture. Both the grazed and ungrazed areas are on a slope. There is an extensive amount of knapweed throughout this site. 96  The grazed area had a lot of quackgrass as well as Kentucky bluegrass. It has a hummocky terrain as a result of glacial activity. This site was also next to an esker (long, curvy ridge formed during glacial meltdown under ice where streams formçd and debris carried by the glaciers deposited there).  97  Map centre 677252 55551 1 UTM ZiO  Merritt Exciosure sites  Nicola Lake  ‘I.  25  Appendix 7. Map of the Drum Lake, Microwave Repeater, Summit and Goose Lake sites near Merritt, BC  98  Map centre 682329 5629751  Z1O  Lac du Bois Grasslands  Station sites in the Lac du Bois Provincial Park.  I  99  Appendix 8. Map of the Lac du Bois Fertilizer-i, Lac du Bois Fertilizer-6, Deep Lake Fertilizer-i, and Agriculture Canada Weather  Map centre 652343 5608061 UTM Z10  Tunkwa Exciosure sites  Appendix 9. Map of the Tunkwa Lake site near the Tunkwa Lake resort.  100  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0066928/manifest

Comment

Related Items