TESTING INTENSIVE CATTLE GRAZING MANAGEMENT AS A RESTORATION
TOOL IN INVADED, SEMI-ARID GRASSLANDS
by
KRISTI GORDON
Bachelor of Natural Resource Science, Thompson Rivers University, 2017
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF
MASTER OF SCIENCE
in Environmental Science

Thesis examining committee:
Lauchlan Fraser (PhD), Supervisor and Professor in Natural Resource Sciences, Thompson
Rivers University
Wendy Gardner (PhD), Associate Professor and Committee Member in Natural Resource
Sciences, Thompson Rivers University
Thomas Pypker (PhD), Associate Professor and Committee Member, Department of Natural
Resource Sciences, Thompson Rivers University.
Nancy Shackelford (PhD), Assistant Professor and External Examiner, Department of
Environmental Studies, University of Victoria
Thompson Rivers University
© Kristi Gordon, 2022

ABSTRACT
Overgrazing and invasive species pose significant threats to the livelihood of British
Columbia (B.C.) ranchers. Spotted knapweed (Centaurea stoebe), in particular, can reduce
native plant diversity, form dense monocultures and overwhelm the native seed bank.
Integrated land management strategies are therefore needed to suppress weeds and restore
ecological function as a whole. Recent research suggests that short-duration, light to
moderate grazing may aid in restoring grasslands. My primary research objective was to test
management-intensive grazing (MiG), extensive grazing, and targeted cattle grazing for their
effects on: a) plant community structure and productivity, b) soil chemical and physical
properties, and c) the storage of soil carbon in a semi-arid grassland. A secondary objective
was to determine the efficacy of targeted cattle grazing to suppress C. stoebe seed
production. Electric fence enclosures were established in C. stoebe dominated grassland in
Merritt, BC. Cattle numbers and timing were controlled such that MiG was ten cow/calf pairs
for one day at the end of the summer growing season, extensive was one cow/calf pair for ten
days at the end of the summer growing season and targeted was ten cow/calf pairs for one
day at the height of spotted knapweed flowering. Results demonstrated that MiG improved
native grass cover and productivity, but total productivity, diversity indices or soil properties
did not differ from extensive grazing. Targeted cattle grazing was effective in controlling C.
stoebe seed production; cattle readily consumed C. stoebe at the late bud-flowering stage and
reduced the number of mature seeds by 88% and seed heads by 79%. At the point of targeted
grazing, C. stoebe also contained more crude protein and total digestible nutrients than the
grass community. This research helped demonstrate that intensive grazing practices have the
potential to create productive invasive-free grasslands in B.C.’s southern interior and beyond.
Research results generated recommendations for implementing MiG on natural semi-arid
grasslands, as well as for targeted cattle grazing for C. stoebe control.

Key words: Grasslands, spotted knapweed, management-intensive grazing, soil carbon
sequestration, targeted grazing, forage quality, soil seed bank, biodiversity

ii

TABLE OF CONTENTS
ABSTRACT ............................................................................................................................................................ II
LIST OF TABLES ................................................................................................................................................... VI
LIST OF FIGURES ................................................................................................................................................ VII
ACKNOWLEDGEMENTS ...................................................................................................................................... IX
1.

CHAPTER 1 ..................................................................................................................................... 10
INTRODUCTION: GRASSLANDS AND RESTORATION......................................................................................... 10
GRASSLANDS AND THEIR IMPORTANCE ...................................................................................................................... 10
GRASSLANDS ARE THREATENED................................................................................................................................ 11
Overgrazing ................................................................................................................................................. 11
Climate Change ........................................................................................................................................... 12
Invasive Species ........................................................................................................................................... 12
RANGELAND RESTORATION ..................................................................................................................................... 14
GRAZING AS A RESTORATION TOOL........................................................................................................................... 15
Extensive Grazing ........................................................................................................................................ 16
Management-Intensive Grazing.................................................................................................................. 17
Targeted Grazing ......................................................................................................................................... 18
KNOWLEDGE GAPS & THESIS OBJECTIVES .................................................................................................................. 18
REFERENCES ..................................................................................................................................................... 20

2.

CHAPTER 2 ..................................................................................................................................... 26
TESTING INTENSIVE CATTLE GRAZING MANAGEMENT TO RESTORE INVADED, SEMI-ARID RANGELANDS ..... 26
GENERAL INTRODUCTION........................................................................................................................................ 26
Management-intensive Grazing .................................................................................................................. 26
Extensive Grazing ........................................................................................................................................ 27
Targeted Grazing ......................................................................................................................................... 28
ENVIRONMENTAL RESPONSES TO GRAZING MANAGEMENT ........................................................................................... 28
Vegetation Responses ................................................................................................................................. 28
Soil Responses.............................................................................................................................................. 29
Particulate and Mineral-Associated Organic Matter and Soil Fractionation .............................................. 29
RESEARCH OBJECTIVES ........................................................................................................................................... 30
MATERIALS AND METHODS ..................................................................................................................................... 32
EXPERIMENTAL DESIGN (FIELD GRAZING TRIALS) ........................................................................................................ 33
Ethics Statement.......................................................................................................................................... 34

iii

VEGETATION SAMPLING ......................................................................................................................................... 34
Plant Community Composition & Diversity ................................................................................................. 34
Biomass ....................................................................................................................................................... 35
Forage Nutritive Values ............................................................................................................................... 35
SOIL SAMPLING..................................................................................................................................................... 36
Bulk Density ................................................................................................................................................. 36
Soil pH .......................................................................................................................................................... 37
Soil Organic Matter (SOM) .......................................................................................................................... 37
Soil Particulate Fractionation ...................................................................................................................... 37
STATISTICAL ANALYSIS ............................................................................................................................................ 39
RESULTS .............................................................................................................................................................. 40
Plant Community Structure ......................................................................................................................... 40
Diversity Indices ........................................................................................................................................... 43
Biomass ....................................................................................................................................................... 44
Forage Nutritive Values ............................................................................................................................... 46
Soil Bulk Density .......................................................................................................................................... 48
Soil pH .......................................................................................................................................................... 48
Soil Organic Matter (SOM) .......................................................................................................................... 49
Soil Fractionation......................................................................................................................................... 50
DISCUSSION ...................................................................................................................................................... 53
Plant Community Composition and Diversity .............................................................................................. 53
Biomass ....................................................................................................................................................... 56
Forage Nutritive Values ............................................................................................................................... 57
Targeted Cattle Grazing and Spotted Knapweed ........................................................................................ 58
Soil Responses to Grazing Management ..................................................................................................... 59
CONCLUSIONS & FUTURE RESEARCH......................................................................................................................... 61
REFERENCES ..................................................................................................................................................... 64
3.

CHAPTER 3 ..................................................................................................................................... 73
TESTING THE EFFICACY OF TARGETED CATTLE GRAZING TO SUPPRESS SPOTTED KNAPWEED (CENTAUREA
STOEBE) IN SEMI-ARID RANGELANDS .............................................................................................................. 73
INTRODUCTION ..................................................................................................................................................... 73
RESEARCH OBJECTIVES ........................................................................................................................................... 76
MATERIALS AND METHODS ..................................................................................................................................... 77
Field Experiment .......................................................................................................................................... 77
ETHICS STATEMENT ............................................................................................................................................... 78

iv

SPOTTED KNAPWEED PLANT TRAITS & RESPONSE TO TARGETED GRAZING....................................................................... 78
Soil Seed Bank Sampling .............................................................................................................................. 79
GREENHOUSE STUDY ............................................................................................................................................. 80
STATISTICAL ANALYSIS ............................................................................................................................................ 81
RESULTS .............................................................................................................................................................. 82
Knapweed Response to Targeted Grazing .................................................................................................. 82
Soil Seed Bank.............................................................................................................................................. 83
DISCUSSION ......................................................................................................................................................... 85
Soil Seed Bank.............................................................................................................................................. 87
CONCLUSIONS & THE FUTURE OF TARGETING GRAZING ............................................................................................... 88
REFERENCES ..................................................................................................................................................... 90
4.

CHAPTER 4 ..................................................................................................................................... 94
RESEARCH SUMMARY AND MANAGEMENT IMPLICATIONS ............................................................................ 94
RESEARCH SUMMARY AND RECOMMENDATIONS......................................................................................................... 94
FURTHER RESEARCH OPPORTUNITIES ........................................................................................................................ 96
REFERENCES ..................................................................................................................................................... 98
APPENDIX 1 – VEGETATION SURVEY PLANT LIST FROM DRUM LAKE PASTURE IN MERRITT, B.C. ......................................... 100
APPENDIX 2 – THE ‘PAR + DEN 5’ METHOD DESCRIBED IN THE MODIFIED FRACTIONATION PROTOCOL RECEIVED FROM THE
SUMMERLAND RESEARCH AND DEVELOPMENT CENTRE (SURDC), ONE OF AGRICULTURE AND AGRI-FOOD CANADA’S RESEARCH
CENTRES. ........................................................................................................................................................... 101

APPENDIX 3 – SUMMARIZED VEGETATION SURVEY DATA FOR THE TARGETED AND CONTROL TREATMENT ENCLOSURES AT DRUM
LAKE PASTURE. ................................................................................................................................................... 106

v

LIST OF TABLES
TABLE 2.1. SUMMARIZED RAINFALL (MM) VALUES FOR EACH MONTH OF THE GROWING SEASON (MAY TO SEPTEMBER) IN 2019 AND
2020. DATA RETRIEVED FROM WEATHER STATION IDENTIFIER # 1125073, LOCATED IN MERRITT, BRITISH COLUMBIA (ECCC
2021). ........................................................................................................................................................... 33
TABLE 2.2. DESCRIPTION OF THE GRAZING SCHEDULE FOR DRUM LAKE PASTURE AS OUTLINED IN THE 2017 - 2021 RUP PREPARED
BY CHUTTER RANCH LTD. ................................................................................................................................... 33

TABLE 2.3. SUMMARIZED LINEAR MIXED EFFECT MODELS FOR THE VEGETATION COMPOSITION AND GROUND COVER RESPONSE
VARIABLES, INCLUDING THE TRANSFORMED ESTIMATED MEANS AND STANDARD ERROR VALUES . SIGNIFICANT CONTRASTS (Α =

.05) BETWEEN TREATMENTS ARE PROVIDED FROM TUKEY HSD POST-HOC TESTS.......................................................... 41
TABLE 2.4. SUMMARIZED LINEAR MIXED EFFECT MODELS FOR THE SHANNON (H’) AND SIMPSON (D) DIVERSITY INDEX RESPONSE
VARIABLES, INCLUDING THE TRANSFORMED ESTIMATED MEANS AND STANDARD ERROR VALUES . SIGNIFICANT CONTRASTS (Α =

.05) BETWEEN TREATMENTS ARE PROVIDED FROM TUKEY HSD POST-HOC TESTS.......................................................... 43
TABLE 2.5. SUMMARIZED LINEAR MIXED EFFECT MODEL RESULTS FOR THE BIOMASS DATA , INCLUDING THE TRANSFORMED ESTIMATED
MEANS AND STANDARD ERROR VALUES. SIGNIFICANT CONTRASTS (Α = .05) BETWEEN TREATMENTS ARE PROVIDED FROM

TUKEY HSD POST-HOC TESTS.............................................................................................................................. 44
TABLE 2.6. DESCRIPTIVE SUMMARY OF MEAN FORAGE NUTRITIVE VALUES (± STDV.) OF GRAMINOIDS IN LATE JULY (POINT OF
TARGETED GRAZING TREATMENT ) AND EARLY SEPTEMBER, AND SPOTTED KNAPWEED IN LATE JULY................................... 47

TABLE 2.7. SUMMARIZED LINEAR MIXED EFFECT MODELS FOR THE SOIL P H AND SOIL ORGANIC MATTER DATA AT BOTH SAMPLING
DEPTHS (0-10 AND 10-20 CM), INCLUDING THE TRANSFORMED ESTIMATED MEANS AND STANDARD ERROR VALUES .

SIGNIFICANT CONTRASTS (Α = .05) BETWEEN TREATMENTS ARE PROVIDED FROM TUKEY HSD POST-HOC TESTS. ................ 48
TABLE 2.8. SUMMARIZED LINEAR MIXED EFFECT MODEL THE SOIL CARBON , NITROGEN AND C-N RATIO FOR THE PARTICULATE
ORGANIC MATTER (POM) AND MINERAL ASSOCIATED ORGANIC MATTER (MAOM) BY TREATMENT AND SOIL SAMPLING DEPTH

(0-10 OR 10-20 CM). SIGNIFICANT CONTRASTS (Α = .05) BETWEEN TREATMENTS ARE PROVIDED FROM TUKEY HSD POSTHOC TESTS....................................................................................................................................................... 51

TABLE 3.1. DESCRIPTION OF THE GRAZING SCHEDULE FOR DRUM LAKE PASTURE AS OUTLINED IN THE 2017 - 2021 RUP PREPARED
BY CHUTTER RANCH LTD. ................................................................................................................................... 77

TABLE 3.2. SUMMARIZED MEAN SPOTTED KNAPWEED PLANT TRAIT VALUES (+/- 1 SE), IN ADDITION TO THE MEAN KNAPWEED SEEDS
(+/- 1 SE) GROUPED BY SEED MATURITY. MANN-WHITNEY U AND INDEPENDENT SAMPLE T-TEST STATISTICS ARE DISPLAYED.
BOLD P-VALUES DENOTE SIGNIFICANCE BETWEEN THE CONTROL AND TARGETED KNAPWEED PLANT VARIABLES (Α = .05)...... 82
TABLE 3.3. SOIL SEED BANK COMPOSITION WITHIN THE TARGETED AND CONTROL ENCLOSURES AT DRUM LAKE PASTURE. SPECIES
STATUS, THE TOTAL NUMBER OF OBSERVED SEEDLINGS , AND THE AVERAGE SEED DENSITY (M2) FOR EACH SPECIES IS PRESENTED.

..................................................................................................................................................................... 84

vi

LIST OF FIGURES
FIGURE 1.1. THE GRAZING MANAGEMENT CONTINUUM, AS ILLUSTRATED BY ALLEN DOBB (2013). .......................................... 16
FIGURE 1.2. A CONTINUOUS GRAZING SYSTEM WITH NO ROTATION OR FENCING DEVELOPMENTS (LEFT), BASIC GRAZING ROTATIONS
(MIDDLE), AND INTENSIVE ROTATIONAL GRAZING SYSTEMS (RIGHT). PHOTO COURTESY OF THE FOOTHILLS FORAGE & GRAZING
ASSOC. ........................................................................................................................................................... 17
FIGURE 2.1. LOOKING SOUTH-EAST TOWARDS DRUM LAKE IN LATE JUNE. ............................................................................ 32
FIGURE 2.2. MAP DEPICTING THE GRAZING TREATMENT ENCLOSURES , AT DRUM LAKE PASTURE IN MERRITT, BC. ....................... 34
FIGURE 2.3. SOIL SAMPLING GRID WITHIN EACH TREATMENT ENCLOSURE DEPICTING MAIN SOIL SAMPLING POINTS AND THE BULK
DENSITY SAMPLING POINTS. ................................................................................................................................ 36

FIGURE 2.4. MODIFIED SOIL FRACTION PROCEDURE BASED ON THE PAR + DEN5 METHOD DESCRIBED IN POEPLAU ET AL. 2018. .... 38
FIGURE 2.5. VIOLIN PLOTS ILLUSTRATING THE UNTRANSFORMED (A) TOTAL GRASS, (B) NATIVE GRASS, (C) TOTAL FORB, (D) NATIVE
FORB, AND (E) BARE GROUND PERCENT COVER DATA . COMPARISONS ARE MADE BETWEEN THE PRE -TREATMENT (CONTROL)
DATA AND THE THREE GRAZING TREATMENTS AFTER TWO YEARS OF TREATMENTS. DIFFERENT LETTERS BETWEEN TREATMENT
DENOTES SIGNIFICANCE (Α = .05) FROM TUKEY HSD POST-HOC TESTS. ...................................................................... 42

FIGURE 2.6. VIOLIN PLOTS ILLUSTRATING THE UNTRANSFORMED (A) SHANNON (H) DIVERSITY AND, (B) SIMPSON (D) DIVERSITY.
COMPARISONS ARE MADE BETWEEN THE PRE-TREATMENT (CONTROL) DATA AND THE THREE GRAZING TREATMENTS AFTER TWO
YEARS OF TREATMENT. DIFFERENT LETTERS BETWEEN TREATMENT DENOTES SIGNIFICANCE (Α = .05) FROM TUKEY HSD POSTHOC TESTS....................................................................................................................................................... 43

FIGURE 2.7.VIOLIN PLOTS ILLUSTRATING THE UNTRANSFORMED (A) MIXED GRASS, (B) MIXED FORB, (C) SPOTTED KNAPWEED (SKW),
(D) TOTAL LIVE, AND (E) LITTER BIOMASS EXPRESSED AS G/.25M2 IN LATE JULY. DIFFERENT LETTERS BETWEEN TREATMENT
DENOTES SIGNIFICANCE (Α = .05) FROM TUKEY HSD POST-HOC TESTS. ...................................................................... 45

FIGURE 2.8.VIOLIN PLOTS ILLUSTRATING SOIL PH AT THE (A) 0-10 CM AND (B) 10-20 CM SAMPLING DEPTHS. DIFFERENT LETTERS
BETWEEN TREATMENT DENOTES SIGNIFICANCE (Α = .05) FROM TUKEY HSD POST-HOC TESTS. ........................................ 48

FIGURE 2.9. VIOLIN PLOTS ILLUSTRATING THE PERCENT SOIL ORGANIC MATTER DATA AT THE (A) 0-10 CM AND (B) 10-20 CM
SAMPLING DEPTHS. DIFFERENT LETTERS BETWEEN TREATMENT DENOTES SIGNIFICANCE (Α = .05) FROM TUKEY HSD POST-HOC
TESTS. ............................................................................................................................................................ 49

FIGURE 2.10. VIOLIN PLOTS ILLUSTRATING THE (A) % CARBON AND (B) % NITROGEN CONTENT WITHIN THE MINERAL ASSOCIATED
ORGANIC MATTER (MAOM) AT THE 0-10 CM SAMPLING DEPTH INTERVAL. ................................................................ 52

FIGURE 2.11.A LANDSCAPE VIEW OF THE TARGETED GRAZING TREATMENT ENCLOSURE BEFORE (A) AND AFTER (B) GRAZING AT THE
LATE BUD TO FLOWERING STAGE IN EARLY AUGUST, 2020. ...................................................................................... 55

FIGURE 3.1. COMPARISON OF SPOTTED KNAPWEED'S TAPROOT AND THE FIBROUS ROOT STRUCTURE OF BLUEBUNCH WHEATGRASS
(PHOTO CREDITS: MARCI HESS AND SAGE GROUSE INITIATIVE)................................................................................. 73
FIGURE 3.2. MAP ILLUSTRATING THE TARGETED (GRAZED) AND CONTROL (UNGRAZED) TREATMENT ENCLOSURES AND KNAPWEED
SAMPLING TRANSECTS LOCATED AT DRUM LAKE PASTURE IN MERRITT, BC. ................................................................. 77

vii

FIGURE 3.3. A) DISSECTING SPOTTED KNAPWEED SEED HEADS AND SORTING THE KNAPWEED SEEDS BY STAGES OF MATURITY . B) A
CLOSER LOOK AT THE MATURE SPOTTED KNAPWEED SEEDS COLLECTED . ....................................................................... 79

FIGURE 3.4. SOIL SEED BANK SAMPLING DESIGN WITHIN EACH: A) 50 M BY 50 M TREATMENT ENCLOSURE, B) 1M2 SAMPLING
QUADRAT FRAME.............................................................................................................................................. 80

FIGURE 3.5. THE SOIL SEED BANK GERMINATION TRAYS SET UP IN THE GREENHOUSE , AND A SELECTION OF SEEDLINGS IDENTIFIED IN
THE SEED GERMINATION EXPERIMENT : A) MICROSTERIS GRACILIS, B) CENTAUREA STOEBE ROSETTE, C) COLLINSIA PARVIFLORA
AND D) LUPINUS SERICEUS. ................................................................................................................................ 81

FIGURE 3.6. SPOTTED KNAPWEED SEED CACHES, PRESUMABLY FORMED FROM THE SMALL MAMMAL POPULATION ON SITE . .......... 88

viii

ACKNOWLEDGEMENTS
I would like to thank Dr. Lauchlan Fraser for his endless support, mentorship, and inspiration
over the past three years. You encouraged me to reach for my goals when times were tough,
and to believe in myself as a scientist.
Dr. Wendy Gardner and Dr. Tom Pypker for agreeing to be on my committee, providing
thoughtful feedback on my writing, and cheering me on along the way. Dr. Nancy
Shackelford for being the external examiner and sharing her expertise in ecological
restoration practices and statistical analysis.
Dr. Andrew Midwood from UBCO for inviting me to learn more about the soil carbon
fractionation procedure, and for providing answers to my questions.
Marc Nadeau for helping me navigate and troubleshoot the research greenhouse control
system.
Fraser Lab Analytical and Cumberland Valley Analytical Services for conducting the wet
chemistry analyses on forage samples.
I would like to thank Dave Chutter and Chutter Ranch LD for allowing us to access and work
on their crown range tenure and with their cattle. This research wouldn’t be possible without
the support of Mark and Cara Haywood Farmer who assisted with the transportation of cattle
and maintenance of electric fencing; Curt Thompson for installment of electric fencing; BC
Cattlemen’s Association; Grasslands Conservation Council of BC; BC Ministry of Forests,
Lands, Natural Resource Operations and Rural Development; BC Ministry of Transportation;
Natural Sciences and Engineering Research Council; Nicola Watershed Community
Roundtable; Thompson Nicola Invasive Plant Management Committee.
Thank you to everyone at the Fraser Lab for your laughs, encouragement, and support - both
in the field and in the lab. I’ve made memories that will last a lifetime with you all.
Thank you to my family and friends for inspiring me to finish my thesis. Thank you for your
patience while I navigated my research, and for lifting me up whenever I felt discouraged. I
would not have been able to complete my thesis without your love and support.
ix

1. CHAPTER 1
INTRODUCTION: GRASSLANDS AND RESTORATION
Grasslands and Their Importance
Grasslands historically covered 61.5 million hectares of Canadian soil (Clayton et al. 1977),
however, only 11.4 million hectares remain today (Alemu et al. 2019). In the Canadian
prairie provinces, including Alberta, Saskatchewan, and Manitoba, much of the native
grasslands have been transformed into cultivated cropland or hay fields (Vankosky et al.
2017). However, in British Columbia (B.C.), approximately 1% of the native grasslands
remain intact (Iverson 2004) and play important cultural, societal, environmental, and
economic roles.
Indigenous communities have made use of the grassland’s abundant flora for food,
medicinal, and ceremonial purposes for time immemorial (Blackstock and McAllister 2004).
Land users explore and recreate in grasslands for their unique landform features and
aesthetics. Endangered wildlife species spend either a portion or the entirety of their
lifecycles within grasslands (Gayton 2003). Furthermore, organic matter incorporates into the
soil through the annual shedding and decomposition of plant roots; thus, grasslands also
assist in the mitigation of climate change through the uptake and long-term storage of
atmospheric carbon (Gayton 2003; Harrower et al. 2012; Lloyd 2019).
Grasslands are also an important resource for cattle ranching in B.C., an industry estimated to
contribute $600 million to the economy annually (BCCA 2020). Grasslands used for forage
and livestock production are termed ‘rangelands’ (Lund 2007), where approximately 80% of
B.C.’s cattle operations utilize crown - or provincially administered – rangeland (BCCA
2020). Despite forested areas composing 80% of the crown range in B.C. (Wikeem et al.
1993), a small but significant proportion of rangelands are grassland communities and
contain ecologically diverse and threatened ecosystems.

10

Grasslands are Threatened
Improper recreational use, conversion to cropland, and the societal pressures to convert
grasslands into commercial or industrial developments are just a few of the many drivers
causing grassland degradation (Lund 2007; Iverson 2004; Tiscornia et al. 2019). In
particular, B.C.’s livestock and forage producers face challenges in preventing the overuse of
pasture, adapting to the increasingly severe and frequent weather events associated with
climate change, and managing harmful invasive plant species. These three challenges are
discussed below.
Overgrazing
Grasslands can rapidly degrade and take years to recover following continuous and heavy
grazing (Tiscornia et al. 2019). Overgrazing occurs when the forage demand exceeds the
carrying capacity, or the amount of forage available for grazing animals (Teague et al. 2011;
Tiscornia et al. 2019). B.C. interior grasslands historically experienced sporadic, lowintensity grazing by elk, deer, pronghorn, and bighorn sheep (Evans et al. 2012; GCC 2017).
However, the introduction of large domestic cattle herds in the 1830s-50s resulted in heavy,
year-round grazing (Gayton 2003) and rangeland decline. Therefore, this increase in grazing
frequency and intensity placed pressure on a variety of native perennial bunchgrasses,
including bluebunch wheatgrass (Pseudoroegneria spicata), Idaho fescue (Festuca
idahoensis), and rough fescue (Festuca campestris) (Krzic et al. 2014; Meays et al. 2014;
Bradfield et al. 2021).
In addition to altering plant community composition and structure (Rambo and Faeth 1999;
Wan et al. 2011; Li et al. 2016), overgrazing can harm soil health. Bulk density, a measure of
soil compaction, can increase in response to grazing in semi-arid rangeland when compared
to adjacent ungrazed areas (Chanasyk and Naeth 1995, Krzic et al. 2014). Increased soil
compaction makes root penetration difficult and, therefore, reduces water infiltration and the
amount of water available to plants (Donkor et al. 2002; Byrnes et al. 2018). In grasslands,
overgrazing can also expose bare ground, leading to erosion and the loss of essential soil
nutrients (Iverson 2004; Metera et al. 2010; Krzic et al. 2014). Moreover, up to 40% of the

11

carbon stored within the soil surface can be lost in overgrazed areas compared to historically
ungrazed, natural grasslands (Wang et al. 2014).
Climate Change
Climate change is another concern facing B.C.’s agriculture industries. Annual increases in
temperature and atmospheric C02 levels, altered precipitation patterns, and lower soil
moisture values are several anticipated effects of climate change in western grasslands
(Gayton 2013; Densmore-McCulloch et al. 2016; Belovsky and Slade 2020; Finch et al.
2021). These changes are expected to alter the growth, distribution, and reproductive success
of grassland plants (Hamim 2005; Gayon 2013; Finch et al. 2021).
The cool season (C3) bunchgrasses that typically dominate B.C.’s grasslands, such as
bluebunch wheatgrass, are expected to withstand the increased temperatures, as they become
dormant in the drier summer months (Gayton 2013). Seasonal changes may also occur where
grasslands become wetter and cooler during active spring growth (May and June) and drier
and hotter throughout the summer (Belovsky and Slade 2020). These changes, in addition to
increased atmospheric C02 levels, may actually allow C3 grassland species to maintain or
improve their levels of productivity in response to changing climates (Taub 2010).
However, invasive plants are also known to be adaptable and highly plastic to environmental
conditions (Clements and Ditommaso 2010), therefore it is expected that more frequent and
severe weather events will provide opportunities for invasive plants to establish and persist
(Hobbs and Huenneke 1992, Teague et al. 2008).
Invasive Species
A species is considered to be invasive if they are found outside of their natural range and
cause harm to the surrounding environment, to the economy, or to society (ISCBC 2017).
Once invasive plants have established, they can be difficult and costly to remove (Rankin et
al. 2004). In B.C.’s Southern Interior rangelands, invasive plants have reduced forage
production for livestock, and have created vulnerable, less diverse plant communities
(Maxwell et al. 1992; Simberloff et al. 2013; Schuster et al. 2018).

12

Invasive plants can be toxic and unpalatable to livestock, and can therefore reduce forage
quality. Hoary alyssum (Berteroa incana), for instance, is a common rangeland invader in the
mustard (Brassicaceae) family that is toxic when ingested by horses (Madani et al. 2010).
Invasive plants, such as hound’s tongue (Cynoglossum officinale), can also have barbed seeds
(burrs) that ‘hitchhike’ and spread rapidly by travelling on animal fur and clothing (DeClerck Floate 1997). Furthermore, most invasive plants are prolific seed producers and can
quickly overwhelm the native seed bank (Gallandt 2006), many of which can persist in the
soil for years. One rangeland invasive plant in particular, spotted knapweed (Centaurea
stoebe), threatens the livelihood of B.C.’s ranching communities and can degrade B.C.’s
sensitive and endangered grassland ecosystems.
Spotted Knapweed
Spotted knapweed (Centaurea stoebe) is an herbaceous forb introduced from Eurasia that has
invaded over 2.9 million hectares of land in North America (Martin et al. 2014). The plant is
a biennial or short-lived perennial that can produce up to 1000 seeds annually (Davis et. al
1993), form dense monocultures, and can produce allelopathic chemical compounds in the
soil, which in turn can inhibit native plant growth and establishment (Martin et al. 2014).
Knapweed is an economic burden, costing B.C. ranchers an estimated $13 million annually
to account for cattle forage losses (Rankin et al. 2004). The widespread use of chemicals to
control invasive plants is also expensive and likely to decrease in the future with increased
efforts to enforce environmentally and economically sustainable weed management practices
(Joyce et al. 2013). Furthermore, the development of herbicide resistance is a concern in
invasive plants, especially if infestations are repeatedly treated with the same active
ingredient over several years (Kumar et al. 2019; Hulme and Liu 2021). For instance, recent
research found that a population of spotted knapweed in the Kootenay region of B.C.
exhibited resistance to clopyralid at rates 32 times greater than the label recommendation
(Magnin and Hall 2016).
Therefore, integrating multiple weed management techniques in rangelands including
grazing, plant competition, biological control insects, and mowing or hand-pulling may
reduce the use of herbicides and help prevent resilience in invasive plants (Lake and Minteer
13

2017). Thus, there is a need to develop innovative invasive plant management strategies that
focus not only on weed suppression, but ecological function and rangeland productivity as a
whole (Sheley et al. 1996; Wilgen et al. 2001).
Rangeland Restoration
Successful rangeland restoration projects consider the health of above and belowground site
characteristics (Smreciu et al. 2003; Alemu et al. 2019). The international standards
developed by the Society for Ecological Restoration define ecological restoration as “the
process of assisting the recovery of an ecosystem that has been degraded, damaged or
destroyed.” (Gann et al. 2019; Gerwing et al. 2021). Therefore, restoration activities can help
degraded sites regain ecosystem function, such as improved soil nutrients or biomass
production, and ecosystem structure, which includes species diversity and richness (Dobson
et al. 1997; Lima et al. 2016).

Restoration activities on degraded sites typically involve human interventions, such as
removing non-native plant species, revegetating a site, or modifying grazing regimes
(Bradshaw 1996; Hobbs et al. 2011; Shackelford et al. 2013). Ecological succession, or
natural processes (Dobson et al. 1997), can also aid in restoring degraded sites. However,
primary and secondary succession, which includes soil development and the reassembly of
biological communities following either natural or anthropogenic disturbances (Chang and
Turner 2019), may take hundreds to thousands of years (Bradshaw 1996). Therefore, natural
processes will also likely require assistance to be successful in present day restoration
practices, but should be incorporated into restoration plans whenever possible (Bradshaw
1996).
Fortunately, restoration ecology is a rapidly developing field (Aronson et al. 2006;
Shackelford et al. 2013; Martin 2017; Gerwig et al. 2021), and research within native
grassland ecosystems is on the rise (Wang et al. 2014; Prive et al. 2021). Specific research
areas of interest are the re-establishment of native plant communities’ post-disturbance and
the potential for soil carbon sequestration in grasslands (Conant et al. 2001; Harrower et al.
2012; Bork et al. 2020; Prive et al. 2021).
14

Grazing as a Restoration Tool
It is well documented that implementing improved grazing management practices can
promote rangeland health and associated ecosystem services (Conant et al. 2003; Alemu et
al. 2019; Bork et al. 2021). Grasslands have co-evolved with forms of natural disturbances,
such as grazing from ungulates (Teague et al. 2008). Therefore, plant communities may
become may become altered once the disturbance is removed. For instance, removing
grazing entirely in native grasslands can deplete biodiversity; in some cases, over 60% of
species have been lost in grasslands after long-term grazing suppression (Metera et al. 2010).
Moderately grazed grasslands can increase above and belowground carbon levels through
compensatory growth and the turnover of plant roots (Schönbach et al. 2011). Hoof action, or
treading by cattle can create small openings in the soil surface that can help native grasses
establish (Savory and Parsons 1980; Metera et al. 2010). While grazers can remove and
ingest large quantities of biomass on the landscape, they can also return carbon in the form of
dung and nitrogen through urine (Whitehead 2020). Grazing animals can also increase the
availability of forage nutrients through excrement (Teague et al. 2008), which are critical for
livestock nutrition and the production of milk and muscle development (Zhai et al. 2018).
Therefore, incorporating cattle grazing may be considered a beneficial tool to help restore
native rangelands (Coffey 2007, GCC 2017).
Grazing management, as defined by Allen Dobb (2013), is the “manipulation of grazing to
achieve an objective or a set of objectives”. Management practices are broad, and ranchers
adopt grazing techniques based on their goals, landscape type, climatic conditions, and
available resources (Teague et al. 2011; Heiberg and Syse 2020). Grazing management can
be viewed along a “continuum” that ranges from extensive grazing (conventional, longduration, high frequency) to management-intensive grazing (short-duration, high intensitylow frequency, adaptive multi-paddock grazing) or MiG (Dobb 2013; Denesiuk 2015;
Heiberg and Syse 2020) (Figure 1.1). As a result, the labour, energy and resources required
to perform MiG practices are much greater than extensive (Figure 1.1). Approximately 66%
of Canadian beef producers currently use a form of extensive grazing management, whereas

15

the remaining 34% utilize MiG (Alemu et al. 2019). However, the percentage of native
grasslands being used for MiG and extensive grazing practices is unknown.

Figure 1.1. The grazing management continuum, as illustrated by Allen Dobb (2013).
Extensive Grazing
Extensive (conventional, rotational) grazing practices are characterized by large areas of land
per animal for long grazing durations and require low manual labor and capital inputs (Dobb
2013) (Figure 1.2). This form of management enables cattle to influence the composition of
vegetation and dominant plant species over time, as highly selected for, or palatable plants
have the opportunity to be repeatedly grazed (Briske et al. 2008; Metera et al. 2010).
Increased frequency and intensity of grazing can also lead to high levels of stress on forage
plants, and eventually, cause mortality (Teague et al. 2008). Therefore, repeatedly grazed
systems often experience a decrease in palatable forage plants and an increase in unpalatable
or undesired plant species as time progresses (Baranova et al. 2019).

16

Ranch operations that have adopted intensive management instead of extensive have
demonstrated an array of benefits in agronomic, irrigated, or seeded pasture systems (Conant
et al. 2003; Teague et al. 2008; Teague et al. 2013).

Figure 1.2. A continuous grazing system with no rotation or fencing developments (left),
basic grazing rotations (middle), and intensive rotational grazing systems (right). Photo
courtesy of the Foothills Forage & Grazing Assoc.
Management-Intensive Grazing
Management-intensive grazing (MiG), also known as adaptive paddock management (ADP)
(Bork et al. 2021), involves grazing many small paddocks at high stocking densities, then
moving the animals into fresh rested pasture daily, often through the use of electric fencing
(Savory and Parsons 1980; Conant et al. 2003; Dobb 2013; Bork et al. 2021) (Figure 1.2).
MiG practices help promote uniform forage use and provide forage species with adequate
time for rest and recovery (Conant et al. 2003).
Proper grazing utilization rates are critical when practicing MiG; ideally, at least 50% or
more of the plant material (by weight) should remain post-grazing (Gerrish 2005; Kyle
2015). Residual plant materials increase photosynthetic capabilities and create enhanced
opportunities for regrowth (Teague et al. 2008), whereas a utilization rate greater than 50%
can result in suppressed above and belowground plant growth (Briske et al. 2008).
Additionally, MiG practices have been shown to reduce erosion through improved root
structures (Reeder and Schuman 2002; Teague et al. 2011; Teague et al. 2013). Teague et al.
(2008) claims that MiG practices have little to no negative effects on soil aggregate
formation or compaction. Increases in soil organic carbon when practicing MiG have also
17

been reported on pastureland, or land dedicated to haying or agronomic species, such as
orchard grass (Dactylis glomerata), Kentucky bluegrass (Poa pratensis), and white clover
(Triflorum repens) (Conant et al. 2003).
Ideally, we can use MiG to produce resilient rangelands, while balancing livestock
production and the sustainable use of forage (Dobb 2013). However, most research on MiG
has focused on irrigated or seeded pasture systems. Therefore, there is little known on the
effects of MiG practices on native, semi-arid grasslands.
Targeted Grazing
Targeted or prescribed grazing is considered a form of intensive management, although the
objectives shift from livestock production to vegetation management and landscape
enhancement (Coffey 2007). Targeted grazing practices alter the duration, intensity, and
season of grazing to achieve specific vegetation management objectives, such as invasive
plant control (Bailey et al. 2019). This form of management provides commodities for the
rancher, such as beef and forage production, in the process of suppressing invasive plants and
creating a more productive range (Rinella and Bellows 2016). Targeted grazing, therefore,
has the potential to be a restoration tool within invaded rangelands.
Knowledge Gaps & Thesis Objectives
There are several knowledge gaps that my research will attempt to address:
1) Addressing a lack of locally-relevant, applied grazing research in B.C.’s Southern
interior, semi-arid rangelands.
2) Most applications of intensive grazing management are performed on irrigated, or
agronomic (seeded) pastures; will we be able to detect short-term effects of intensive
grazing in semi-arid, native rangeland?
3) Will targeted cattle grazing help suppress the invasive plant spotted knapweed?
The overarching purpose of this study was to assess whether intensive grazing management
practices could aid in restoring invaded, semi-arid rangelands in the Southern Interior of B.C.
18

To do this, we conducted a 3-year controlled field experiment which tested the use of MiG,
targeted, and extensive cattle grazing for their effects on plant community structure,
diversity, productivity and soil properties. This field study is summarized in Chapter 2,
‘Testing Intensive Cattle Grazing Management to Restore Invaded, Semi-Arid
Rangelands’.
A secondary research objective was to test targeted cattle grazing to help control the growth
and spread of spotted knapweed. In the field, we compared the growth and seed production
of spotted knapweed plants that had been targeted by cattle or left ungrazed (control). Soil
seed bank samples were collected post-grazing and grown in the greenhouse to compare
knapweed seedling recruitment. Soil seed bank composition, size, and density were also
described for restoration purposes. This field and greenhouse study is summarized in Chapter
3, ‘Testing Targeted Cattle Grazing to Supress Spotted Knapweed (Centaurea stoebe) in
Semi-Arid Rangelands’
Chapter 4, ‘Research Summary and Conclusions’ considers the management implications
of my findings and provides recommendations for future research, particularly for
implementing targeted grazing on a landscape scale.

19

REFERENCES
Alemu A, Kröbel R, McConkey B, Iwaasa A. 2019. Effect of increasing species diversity and
grazing management on pasture productivity, animal performance, and soil carbon
sequestration of re-established pasture in Canadian prairie. Animals. 9(4):127.
Archer SR, Andersen EM, Predick KI, Schwinning S, Steidl RJ, Woods SR. 2017. Woody
plant encroachment: causes and consequences. In: Briske DD, editor. Rangeland systems:
processes, management and challenges. Cham, Switzerland: Springer Nature. p. 25–84.
Aronson J, Clewell AF, Blignaut JN, Milton SJ. 2006. Ecological restoration: A new frontier
for nature conservation and economics. J Nat Conserv. 14(3–4):135–139.
Bailey DW, Mosley JC, Estell RE, Cibils AF, Horney M, Hendrickson JR, Walker JW,
Launchbaugh KL, Burritt EA. 2019. Synthesis Paper: Targeted livestock grazing:
prescription for healthy rangelands. Rangel Ecol Manag. 72(6):865–877.
Baranova A, Oldeland J, Wang S li, Schickhoff U. 2019. Grazing impact on forage quality
and macronutrient content of rangelands in Qilian Mountains, NW China. J Mt Sci.
16(1):43–53.
Beck PA, Beck MR, Gunter SA, Biermacher JT, Gillen RL. 2020. Stocking rate impacts
performance and economics of grazing beef steers on mixed-grass prairies of the Southern
Great Plains. Transl Anim Sci. 4(3):1–13.
Belovsky GE, Slade JB. 2020. Climate change and primary production: Forty years in a
bunchgrass prairie. PLoS One. 15(12 December):1–21.
Bradshaw AD. 1996. Underlying principles of restoration. Can J Fish Aquat Sci. 53(SUPPL.
1):3–9. doi:10.1139/cjfas-53-s1-3.
British Columbia Cattleman’s Association (BCCA). [Internet]. 2020. BC Beef Cattle
Industry. Kamloops, BC: British Columbia Cattlemen’s Association. Available from:
https://www.cattlemen.bc.ca/industry.htm
[BCMoF] British Columbia Ministry of Forests. 2000. Range use plan guidebook. 2nd ed.
Victoria, BC: Government of British Columbia.
Blackstock M, McAllister R. 2004. First nations perspectives on the grasslands of the Interior
of British Columbia. J Ecol Anthropol. 8(1):24–46.
Bork EW, Raatz LL, Carlyle CN, Hewins DB, Thompson KA. 2020. Soil carbon increases
with long-term cattle stocking in northern temperate grasslands. Soil Use Manag. 36(3):387–
399.
Chanasyk DS, Naeth MA. 1995. Grazing impacts on bulk density and soil strength in the
foothills fescue grasslands of Alberta, Canada. Can J Soil Sci. 75(4):551–557.
20

Clayton JS, Ehrlich WA, Cann DB, Day JH, Marshall IB. 1977. Soils of Canada. Canada
Department of Agriculture, Research Branch: Ottawa, ON. Volume 1, p. 243.
Clements DR, Ditommaso A. 2011. Climate change and weed adaptation: Can evolution of
invasive plants lead to greater range expansion than forecasted? Weed Res. 51(3):227–240.
Coffey L. 2007. Targeted Grazing: A natural approach to vegetation management and
landscape enhancement. Launchbaugh KL, editor. Centennial, CO: American Sheep Industry
Association.
Conant RT, Paustian K, Elliott ET. 2001. Grassland management and conversion into
grassland: effects on soil carbon. Ecol Appl. 11(2):343–355.
Conant RT, Six J, Paustian K. 2003. Land use effects on soil carbon fractions in the
southeastern United States. Biol Fertil Soils. 38(6):386–392.
Davis ES, Fay PK, Chicoine TK, Lacey CA. 1993. Persistence of spotted knapweed
(Centaurea maculosa) seed in soil. Weed Sci. 41(1):57–61.
De Bruijn SL, Bork EW, Grekul CW. 2010. Neighbor defoliation regulates Canada thistle
(Cirsium arvense) in pasture by mediating interspecific competition. Crop Prot. 29(12):1489–
1495.
Denesiuk DR. 2017. Grazing management as a climate change adaptation and mitigation
strategy. Thompson Rivers University.
Densmore-McCulloch JA, Thompson DL, Fraser LH. 2016. Short-term effects of changing
precipitation patterns on shrub-steppe grasslands: Seasonal watering is more important than
frequency of watering events. PLoS One. 11(12).
Dobb, A. 2013. BC farm practices and climate change adaptation: Management-intensive
grazing. BC Agriculture and Food Climate Action Initiative. 22p.
Dobson AP, Bradshaw AD, Baker AJM. 1997. Hopes for the future: restoration ecology and
conservation biology. Science (80- ). 277(5325):515–522.
Donkor NT, Gedir J V., Hudson RJ, Bork EW, Chanasyk DS, Naeth MA. 2002. Impacts of
grazing systems on soil compaction and pasture production in Alberta. Can J Soil Sci.
82(1):1–8.
Evans CRW, Krzic M, Broersma K, Thompson DJ. 2012. Long-term grazing effects on
grassland soil properties in southern British Columbia. Can J Soil Sci. 92(4):685–693.
Fraser LH, Carlyle CN. 2011. Is spotted knapweed (Centaurea stoebe L.) patch size related
to the effect on soil and vegetation properties? Plant Ecol. 212(6):975–983.
Gallandt ER. 2006. How can we target the weed seedbank? Weed Sci. 54(3):588–596.

21

Gann GD, McDonald T, Walder B, Aronson J, Nelson CR, Jonson J, Hallett JG, Eisenberg
C, Guariguata MR, Liu J (2019) International principles and standards for the practice of
ecological restoration. Second edition. Restoration Ecology 27:S1–S46
Gayton DV. 2003. British Columbia grasslands: Monitoring vegetation change. Kamloops,
BC: FORREX–Forest Research Extension Partnership. FORREX Series 7. 49p.
Gerwing TG, Hawkes VC, Gann GD, Murphy SD. 2021. Restoration, reclamation, and
rehabilitation: on the need for, and positing a definition of, ecological reclamation. Restor
Ecol.:2–5.
Grasslands Conservation Council of British Columbia (GCC). (2017). Managing BC
grasslands: an overview. Kamloops, BC. 22p.
Harrower W, Carlyle C, DeKoning P, and Fraser L. 2012. Managing rangelands for carbon:
Assessing the viability of carbon offset markets for British Columbia ranchers. 78p.
Heiberg EJ, Syse KL. 2020. Farming autonomy: Canadian beef farmers reclaiming the grass
through management-intensive grazing practices. Org Agric. 10(4):471–486.
Hobbs RJ, Huenneke LF. 1992. Disturbance, diversity, and invasion: Implications for
conservation. Conserv Biol. 6(3):324–337.
Hulme PE, Liu W. 2021. Species prevalence and plant traits discriminate between herbicide
resistant and susceptible weeds. Pest Manag Sci. (August).
Invasive Species Council of British Columbia (ISCBC). 2017. Invasive species strategy for
British Columbia 2017-2022. Williams Lake, BC: Invasive Species Council of BC. 11p.
Iverson K. 2004. Grasslands of the southern interior. Victoria, BC: Ministry of Sustainable
Resource Management. 6p.
Joyce LA, Briske DD, Brown JR, Polley HW, McCarl BA, Bailey DW. 2013. Climate
change and North American rangelands: assessment of mitigation and adaptation strategies.
Rangel Ecol Manag. 66(5):512–528.
Knochel DG, Monson ND, Seastedt TR. 2010. Additive effects of aboveground and
belowground herbivores on the dominance of spotted knapweed (Centaurea stoebe).
Oecologia. 164(3):701–712.
Krzic M, Lamagna SF, Newman RF, Bradfield G, Wallace BM. 2014. Long-term grazing
effects on rough fescue grassland soils in southern British Columbia. Can J Soil Sci.
94(3):337–345.
Kyle, J. [Internet]. 2015. Take half and leave half: pasture management strategies for
sustainable productivity. Guelph, Ontario: Ontario Ministry of Agriculture, Food and Rural
Affairs; [updated July 6 2015]. Available from:
http://www.omafra.gov.on.ca/english/livestock/beef/news/vbn0715a1.htm
22

Kumar A., Choudhary T., Das S., Meena S.K. 2019. Weed seed bank: impacts and
management for future crop production. In: Hasanuzzaman M. (eds) Agronomic Crops.
Springer, Singapore.
Lake EC, Minteer CR. 2018. A review of the integration of classical biological control with
other techniques to manage invasive weeds in natural areas and rangelands. BioControl.
63(1):71–86. doi:10.1007/s10526-017-9853-5.
Launchbaugh KL, Walker JW, Daines R. 2006. Targeted grazing: a new approach to
vegetation management and landscape enhancement. Centennial, CO: American Sheep
Industry Association 199 pp.
Li W, Xu F, Zheng S, Taube F, Bai Y. 2017. Patterns and thresholds of grazing-induced
changes in community structure and ecosystem functioning: species-level responses and the
critical role of species traits. J Appl Ecol. 54(3):963–975.
Li J, Ramirez GH, Kiani M, Quideau S, Smith E, Janzen H, Larney F, Puurveen D. 2018.
Soil organic matter dynamics in long-term temperate agroecosystems: Rotation and nutrient
addition effects. Can J Soil Sci. 98(2):232–245.
Lima AT, Mitchell K, O’Connell DW, Verhoeven J, Van Cappellen P. 2016. The legacy of
surface mining: Remediation, restoration, reclamation and rehabilitation. Environ Sci Policy.
66:227–233.
Lloyd D, Iverson K, Mackenzie K, Richardson H, Jeffries L. 2019. Laurie Guichon
Memorial Grasslands Interpretive Site: A Baseline Inventory. Kamloops, BC. Grassland
Conservation Council of British Columbia. 98 p.
Lund GH. 2007. Accounting for the world’s rangelands. Rangelands. 29(1):3–10.
Madani H, Stopps G, Upadhyaya MK. 2010. Seed germination ecophysiology of hoary
alyssum (Berteroa incana L.) in different temperature, dry heat and anaerobiosis conditions.
J Plant Ecophysiol. 2(2010):121–126.
Mangin AR, Hall LM. 2016. First report: spotted knapweed (Centaurea stoebe) resistance to
auxinic herbicides. Can J Plant Sci. 96(6):928–931.
Martin DM. 2017. Ecological restoration should be redefined for the twenty-first century.
Restor Ecol. 25(5):668–673.
Martin LM, MacDonald NW, Brown TE. 2014. Native plant establishment success
influenced by spotted knapweed (Centaurea stoebe) control method. Ecol Restor. 32(3):282–
294.
Maxwell JF, Drinkwater R, Clark D, Hall JW. 1992. Effect of grazing, spraying, and seeding
on knapweed in British Columbia. J Range Manag. 45(2):180.

23

Meays CL, Laliberte AS, Doescher PS. 2000. Defoliation response of bluebunch wheatgrass
and crested wheatgrass: why we cannot graze these two species in the same manner.
Rangelands. 22(6):6–8.
Metera E, Sakowski T, Słoniewski K, Romanowicz B. 2010. Grazing as a tool to maintain
biodiversity of grassland - a review. Anim Sci Pap Reports. 28(4):315–334.
Palit R, Gramig G, Dekeyser ES. 2021. Kentucky bluegrass invasion in the northern great
plains and prospective management approaches to mitigate its spread. Plants. 10(4):1–14.
Prive K, Orr MR, Kilkenny FF, Reauter RJ, Prendeville HR. 2021. Phenological variation in
bluebunch wheatgrass (Pseudoroeneria spicata): implications for seed sourcing, harvest, and
restoration. Land. 10(1064):1–10.
Rambo JL, Faeth SH. 1999. Effect of vertebrate grazing on plant and insect community
structure. Conserv Biol. 13(5):1047–1054.
Rapp, R. 2018. Pasture principles from the soil up. Farm Talk Newspaper. Sept 26, 2018.
Rankin C, Booth J, Cannings S. 2004. Invasive alien species framework for BC: identifying
and addressing threats to biodiversity. 96p.
Reeder JD and Schuman GE. 2002. Influence of livestock grazing on C sequestration in
semi-arid mixed-grass and short-grass rangelands. Environmental Pollution 116:457- 463.
Rinella MJ, Bellows SE. 2016. Evidence-targeted grazing benefits to invaded rangelands can
increase over extended time frames. Rangel Ecol Manag. 69(3):169–172.
Savory A, Parsons SD. 1980. The Savory Grazing Method. Rangelands. 2(6):234–237.
Schaub S, Buchmann N, Lüscher A, Finger R. 2020. Economic benefits from plant species
diversity in intensively managed grasslands. Ecol Econ. 168:1-12.
Schuster MJ, Wragg PD, Reich PB. 2018. Using revegetation to suppress invasive plants in
grasslands and forests. J Appl Ecol. 55(5):2362–2373.
Shackelford N, Hobbs RJ, Burgar JM, Erickson TE, Fontaine JB, Laliberté E, Ramalho CE,
Perring MP, Standish RJ. 2013. Primed for change: Developing ecological restoration for the
21st century. Restor Ecol. 21(3):297–304.
Sheley RL, Svejcar T, Maxwell B. 1996. A theoretical framework for developing
successional weed management strategies on rangeland. Weed Technol. 10(4):766–
773.
Simberloff D, Martin JL, Genovesi P, Maris V, Wardle DA, Aronson J, Courchamp F, Galil
B, García-Berthou E, Pascal M, et al. 2013. Impacts of biological invasions: What’s what and
the way forward. Trends Ecol Evol. 28(1):58–66.

24

Smreciu A, Sinton H, Walker D, Bietz J. 2003. Establishing native plant communities. Orey
D, editor. Edmonton, Alberta: Alberta Agriculture, Food and Rural Development. 93p.
Teague R, Provenza F, Kreuter U, Steffens T, Barnes M. 2013. Multi-paddock grazing on
rangelands: Why the perceptual dichotomy between research results and rancher experience?
J Environ Manage. 128:699–717.
Teague WR, Dowhower SL, Baker SA, Haile N, DeLaune PB, Conover DM. 2011. Grazing
management impacts on vegetation, soil biota and soil chemical, physical and hydrological
properties in tall grass prairie. Agric Ecosyst Environ. 141(3–4):310–322.
Teague R, Provenza F, Norton B, Steffens T, Barnes M, Kothmann M, Roath R. 2008.
Benefits of multi-paddock grazing management on rangelands: Limitations of experimental
grazing research and knowledge gaps. Grasslands Ecol Manag Restor.:41–80.
Tiscornia G, Jaurena M, Baethgen W. 2019. Drivers, process, and consequences of native
grassland degradation: Insights from a literature review and a survey in Río de la Plata
grasslands. Agronomy. 9(5):8–12.
Toledo D, Sanderson M, Spaeth K, Hendrickson J, Printz J. 2014. Extent of Kentucky
bluegrass and its effect on native plant species diversity and ecosystem services in the
northern great plains of the United States. Invasive Plant Sci Manag. 7(4):543–552.
Wan H, Bai Y, Schönbach P, Gierus M, Taube F. 2011. Effects of grazing management
system on plant community structure and functioning in a semiarid steppe: Scaling from
species to community. Plant Soil. 340(1):215–226.
Wang X, Vandenbygaart AJ, McConkey BC. 2014. Land management history of Canadian
grasslands and the impact on soil carbon storage. Rangel Ecol Manag. 67(4):333–343.
Whitehead D. 2020. Management of grazed landscapes to increase soil carbon stocks in
temperate, dryland grasslands. Front Sustain Food Syst. 4(October):1–7.
Wikeem BM, Mclean A, Quinton D, Bawtree A. 1993. An overview of the forage resource
and beef production on Crown land in British Columbia. Can J Anim Sci. 73(4):779–794.
doi:10.4141/cjas93-081.
Wilgen B van, Richardson D, Higgins S. 2001. Integrated control of invasive alien plants in
terrestrial ecosystems. L Use Water Resour Res. 5:6.
Zhai X, Zhang Y, Wang K, Chen Q, Li S, Huang D. 2018. Grazing effects on the nutritive
value of dominant species in steppe grasslands of northern China. BMC Ecol. 18(1):1–9.

25

2. CHAPTER 2
TESTING INTENSIVE CATTLE GRAZING MANAGEMENT TO RESTORE
INVADED, SEMI-ARID RANGELANDS
General Introduction
Grasslands provide essential ecosystem goods and services, such as forage production for
livestock, wildlife habitat, soil carbon sequestration, biodiversity, and more (Wikeem and
Wikeem 2004; Dobb 2013; Augustine et al. 2020; Bork et al. 2021). However, grasslands are
sensitive and can rapidly degrade in response to disturbance. It may take years to decades for
a disturbed grassland to return to its previous vegetation community and state of ecological
functioning (Iverson 2004; Tiscornia et al. 2019).
The introduction of non-native, harmful invasive plant species threatens rangeland ecosystem
goods and services (Wikeem and Wikeem 2004; Clements et al. 2007). Spotted knapweed
(Centaurea stoebe), for instance, is an aggressive, drought-tolerant invader that can disrupt
soil stability and outcompete native plant species (Fraser and Carlyle 2011; Gayton and
Miller 2012). Periodic drought can also make rangeland restoration efforts, such as seeding
and revegetation, difficult due to insufficient moisture and competition from drought-adapted
species (Wikeem and Wikeem 2004; Evans et al. 2012; Krzic et al. 2014). Therefore, the
ranching and scientific communities alike have been developing improved grazing
management strategies in an effort to help restore native rangelands following mild to severe
disturbance events (Conant and Paustian 2001; Dobb 2013; Chen et al. 2015; Döbert et al.
2021). Several of these grazing management systems are described below.
Management-intensive Grazing
Management-intensive grazing (MiG), otherwise known as adaptive multi-paddock
management (AMP), involves grazing at moderate to high stocking rates, then frequently
moving the livestock into fresh rested pasture (Savory and Parsons 1980; Dobb 2013; Mann
and Sherren 2018; Bork et al. 2021). In theory, longer periods of rest between defoliation
events - especially during active spring growth – along with a generous stubble height, allow
26

grazed plants to better regrow and recover (Dobb 2013; Augestine et al. 2020; Bork et al.
2021; Döbert et al. 2021). When practicing MiG, livestock also return carbon and nitrogenrich manure and urine back into the system at a more uniform distribution (Gupta et al. 2016;
Marie et al. 2018). Cow manure, in particular, contains high concentrations of nitrogen,
phosphorus, and sulfur, all of which are essential nutrients for plant growth (Raiesi et al.
2006; Almeida et al. 2019; Bader et al. 2021). Increases in plant productivity, or above and
belowground biomass, also encourages soil organic matter (SOM) formation and increases
the potential for soil carbon storage (Lal 2009; Li et al. 2018; Whitehead 2020).
MiG is commonly practiced on agronomic pastureland, or land that consists of maintained
forage crops (Conant et al. 2003; Dobb 2013). These areas are often more accessible for
machinery, and fencing and irrigation developments may already exist or could be installed
more readily than on native rangeland. However, the feasibility of implementing MiG on
semi-arid rangeland hasn’t been investigated thoroughly, nor have the ecosystem benefits
been explored.
Extensive Grazing
Extensive (conventional, rotational) grazing is widely used in the management of native
rangeland pastures, largely because irrigation and fencing or manual labour may be limited
(DiTomaso 2000; Dobb 2013; Heiberg and Syse 2020). Extensive grazing still involves
rotating livestock into rested pastures; however, unlike MiG, the pastures are larger in size
and grazed for longer time periods, resulting in a lower number of animals per unit land area
and less manual labour and capital inputs for the ranching operation (Dobb 2013; Heiberg
and Syse 2020).
Forage selectivity plays a role in extensive grazing management. When provided with a
choice, livestock will select for preferred forages such as perennial bunchgrasses (Metera et
al. 2010; Pauler et al. 2020). Therefore, selectivity and overgrazing can be a concern if there
is an opportunity for livestock to re-graze preferred plants, or if there is limited time for the
plants to recover following the grazing event. Overgrazing limits plant productivity and
suppresses root formation, leading to issues with soil stability, erosion and compaction
(Metera et al. 2010; Krzic et al. 2014). That being said, if plants are provided with sufficient
27

rest between grazing events or rotations, livestock selectivity can encourage the formation of
a ‘mosaic’ landscape, where plant communities display a more heterogenous and diverse
structure (Pavlů et al. 2006; Metera et al. 2010).
Targeted Grazing
Targeted grazing is another intensive land management tool which utilizes grazing animals to
achieve ecological, economic, or other operational goals such as weed suppression and
control, maintenance or enhancement of wildlife habitat, and improved biodiversity (Coffey
2007; Bailey et al. 2019; Rhodes et al. 2021). Targeted grazing has increased in popularity in
recent years because it is cost-effective, can be used in a variety of environments, and can
provide long-lasting results and commodities for the rancher (Diamond et al. 2009; Davy and
Rinella 2017). Areas that are difficult to access (i.e., steep terrain) or are in close proximity to
water can be reached through targeted grazing, and it often provides the opportunity to
reduce or halt the use of herbicides to help control undesired plant species (Davy and Rinella
2017; Bailey et al. 2019). Cattle, for instance, have been used to help reduce fire fuel loads
and help create fire breaks by grazing the highly flammable introduced grass, cheatgrass
(Bromus tectorum) when it is palatable in the spring (Diamond et al. 2009).
Environmental Responses to Grazing Management
Vegetation Responses
Overall, it is well understood that grazing management can alter the structure and function of
plant communities (Hobbs and Huenneke 1992; Li et al. 2017). Native perennial
bunchgrasses such as rough fescue (Festuca scabrella) and bluebunch wheatgrass
(Pseudoroegneria spicata) are sensitive to heavy and repeated grazing, especially during
active spring growth (Meays et al. 2014). Subsequently, these species experience restricted
above and belowground growth and produce fewer viable seeds (Krzic et al. 2014; Meays et
al. 2014; GCC 2017). Furthermore, introduced, drought-tolerant perennials such as Kentucky
bluegrass (Poa pratensis) and crested wheatgrass (Agropyron cristatum) can withstand heavy
grazing pressure (Meays et al. 2014; Palit et al. 2021). Therefore, both species of introduced

28

grass can dominate the slow-growing native species and transform diverse plant communities
into uniform stands when heavily grazed (Palit et al. 2021).
Soil Responses
Soils are also an integral component within grassland ecosystems and can be influenced by
changes in grazing management. Season-long or continuous grazing can increase bulk
density and disrupt soil aggregates (Donkor et al. 2002), making it more difficult for plant
roots to penetrate the soil and access essential water and nutrients (Lee et al. 2014; Tiscornia
et al. 2019; Shrestha et al. 2020). Furthermore, when managed accordingly, grasslands can
act as a carbon sink and, in turn, promote rangeland restoration (Scurlock and Hall 1998;
Tiscornia et al. 2019; Shrestha et al. 2020). In grasslands, approximately 10% of organic
carbon is held within aboveground biomass, whereas the remaining 90% is stored within the
soil organic matter (SOM) (Reeder and Schuman 2002). SOM is often created through the
deposition of plant litter and plays a significant role in improving soil chemical and physical
properties, such as nutrient storage and retention, water holding capacity, and aggregation
(Conant et al. 2001; Smith et al. 2015; Lavallee et al. 2020). The removal of plant litter and
SOM can also suppress microbial activity (Mohammadi et al. 2011; Witzgall et al. 2021) and
slow the decomposition of organic matter, which subsequently alters the nutrients available
for plants and opportunity for soil carbon storage (Raiesi et al. 2006; Whitehead 2020).
Particulate and Mineral-Associated Organic Matter and Soil Fractionation
To better understand the potential for long-term soil carbon storage and changes in SOM
formation within semi-arid rangelands, isolating the particulate organic matter (POM) from
the mineral-associated organic matter (MAOM) within bulk soil is recommended (Li et al.
2018; Popleau et al. 2018; Lavallee et al. 2020). Particulate organic matter is generally
created through the deposition of plant litter and root fragments, and may be modified by soil
biota (Smith et al. 2015; Lavallee et al. 2020), whereas MAOM is made up of singular,
microscopic organic molecules that are tightly bound to silt and clay-sized particles (<
63 µm) (Lavallee et al. 2020). The main difference between POM and MAOM is that POM
often has no protective mechanism or is located with larger soil aggregates (> 63 µm), which
makes POM more susceptible to decomposition and more vulnerable to changes in land
29

management (Lavallee et al. 2020). Additionally, the carbon-nitrogen (C-N ratio) is generally
lower in MAOM, and it contains nutrients and carbon compounds that are readily available
for use by soil microorganisms and plants (Lavallee et al. 2020; Witzgall et al. 2021).
Moreover, MAOM is depicted as the ‘long-term’ soil carbon storage and is estimated to have
a residence time of decades to centuries, whereas POM can persist within the soil for decades
to less than a decade (Poeplau et al. 2018; Lavallee et al. 2020.
Fractionation is a procedure that separates soils into multiple particle sizes, that helps to
understand and recognize how carbon forms, persists, and functions within the soil itself (de
Moraes et al. 2017; Hewins et al. 2018; Popleau et al. 2018; Lavallee et al. 2020). An
assessment of particulate and mineral-associated organic matter may provide insight into
how carbon is stored in response to rapid changes in land management, more specifically the
alteration of grazing practices in semi-arid rangelands (Conant et al. 2003; de Moraes et al.
2017). Furthermore, investigating the influence of grazing management on soil carbon
sequestration may aid in developing climate change mitigation strategies.
Research Objectives
A three-year controlled field experiment tested the use of MiG, extensive, and targeted cattle
grazing for their effects on plant community structure, diversity, productivity and soil
properties in B.C.’s southern interior rangelands.

The hypothesis was that intensive grazing management practices would improve the
condition of B.C.’s southern interior rangelands, such that intensively-grazed pastures would
demonstrate increased native plant cover, productivity, and improved diversity indices
compared to extensive grazing. Whereas extensive grazing would exhibit more bare ground,
less plant litter, and greater cover and productivity of introduced species than the intensivelygrazed pastures. Additionally, spotted knapweed cover and productivity would decrease in
response to the targeted grazing treatment. Finally, the extensively grazed pasture would
exhibit greater bulk density and less soil organic matter, carbon, and nitrogen when
compared to intensively grazed pastures.

30

Forage nutritive values are also reported in this chapter to describe the forage quality of the
spotted knapweed at the point of the targeted grazing treatment (late-July) versus the mixed
grass community in late-July and mid-September.

31

Materials and Methods
Research plots were located in Drum Lake pasture, east of Merritt, and approximately 3 km
north of the Laurie Guichon Memorial Grasslands Interpretative Site (50° 5'41.89"N,
120°40'25.72"W; elevation: 1052-1065m) (Figure 2.1). Drum Lake pasture is located within
the grassland phase of the Thompson very dry, hot Interior Douglas-fir (IDFxh2a)
biogeoclimatic zone (Newman et al. 2011). The mean annual precipitation is 350-400mm,
35% of it falling as snow (Lloyd et al. 2019). Precipitation data for the 2019 and 2020
growing seasons (May 1st – Sept 30th) were retrieved from an Environment and Climate
Change Canada weather station (Table 2.1). The vegetation community of the IDFxh2a zone
is dominated by bluebunch wheatgrass (Pseudoroegneria spicata), rough fescue (Festuca
campestris), Kentucky bluegrass (Poa pratensis), and silky lupine (Lupinus sericeus) (Lloyd
et al. 1990).

Figure 2.1. Looking south-east towards Drum Lake in late June.
Chutter Ranch Ltd. operates under a range tenure where their approximately 500 cow/calf
pairs graze Drum Lake following a spring/fall grazing rotation. In odd years cattle graze for
10 days in early June and in even years cattle graze for 10 days in mid-September (Table
2.2). The Chutter Ranch Ltd. 2017-2021 Range Use Plan (RUP) set the target browse
utilization rate at 25% of the annual plant growth.
32

Table 2.1. Summarized rainfall (mm) values for each month of the growing season (May to
September) in 2019 and 2020. Data retrieved from weather station identifier # 1125073,
located in Merritt, British Columbia (ECCC 2021).
Total Rainfall (mm)
Month

2019

2020

May

42.5

59.0

June

50.0

47.6

July

29.4

18.4

August

12.8

5.0

September

59.0

7.0

Total

193.7

137

Table 2.2. Description of the grazing schedule for Drum Lake pasture as outlined in the 2017
- 2021 RUP prepared by Chutter Ranch Ltd.

Even years

Odd years

Pasture name

Drum Lake

Drum Lake

Period of use
No. and class of livestock

Sept 19 - 28

June 1 - 10

500 C/C*

500 C/C*
20 Bulls

*C/C = cow-calf pairs

Experimental Design (Field Grazing Trials)
In May 2018, three 50 m by 50 m electric fence treatment enclosures were established
(Figure 2.2). Stocking rate was 0.8 animal unit months (AUM) ha-1 for the grazing
treatments, where cattle numbers and timing were controlled such that MiG was ten cow/calf
(C/C) pairs for one day at the end of the summer growing season (mid-September), extensive
was one C/C pair for ten days at the end of the summer growing season, and targeted was ten
C/C pairs for one day at the height of spotted knapweed flowering (late July-early August).
Baseline (pre-treatment) plant cover data was combined among all treatment enclosures; the
baseline cover data is referred to as the 'control' in this experiment. No baseline data was

33

collected for the soil or biomass data, therefore, biomass and soil data collected in 2019 and
2020 were combined by grazing treatment.
Grazing trials commenced in July 2018 and continued annually for three years (2018-2020).
Ethics Statement
This research was carried out under Thompson Rivers University Animal Care and Use
Protocol No. 101899.

50 m

Figure 2.2. Map depicting the grazing treatment enclosures, at Drum Lake pasture in
Merritt, BC.
Vegetation Sampling
Plant Community Composition & Diversity
Vegetation surveys were conducted annually in late June where species composition was
estimated within each treatment enclosure. Ten 0.25 m2 quadrats were placed by generating
random numbers within a 50 m x 50 m sampling grid (Random.org). Percent canopy cover
was used to measure plant species and ground parameters including bare ground, litter, rock,
cryptogamic crust and dung. Percent cover values were allowed to exceed 100%. A two34

meter buffer zone existed between all sampling quadrats and the electric fencing to reduce
edge effects.
Plant species were assigned into the following categories using the electronic database, EFlora B.C. (Klinkenberg 2021): native grasses, introduced grasses, native forbs, and invasive
forbs in an effort to better understand plant community responses to the grazing treatments.
Shannon (H’) and Simpson (D) diversity indices were calculated with species cover data for
each treatment.
Biomass
Biomass clippings were collected prior to grazing treatments in late-July and mid-September
of 2019 and 2020 to determine available forage and forage quality at the point of grazing.
Biomass was harvested within ten randomly placed 0.25 m2 quadrat frames to ground height,
then sorted into the following categories: mixed grasses, mixed forbs, spotted knapweed, and
litter (standing and ground). Immediately (24-48 hours) following grazing, biomass clipping
was repeated to estimate the biomass utilized by cattle. Biomass utilization was estimated by
calculating the difference between the pre and post grazing total biomass (excluding litter).
Samples were oven-dried at 65°C for 48 hours, then weighed on an analytical balance in
preparation for forage nutritive analysis.
Forage Nutritive Values
Wet chemistry analyses were conducted by Fraser Valley Analytical services LTD in
Abbotsford, B.C. The nutritive values of the mixed grass community at two time periods,
which coincided with the intensive grazing trials (late July and early September), and spotted
knapweed at the point of targeted grazing (late-July) were measured. Six biomass subsamples
were selected from each category based on the largest sample quantities (by weight) and
were packaged in plastic Ziploc bags. The following parameters were measured: acid
detergent fiber (ADF), neutral detergent fiber (NDF), total digestible nutrients (TDN), crude
protein (CP), calcium, phosphorus, magnesium, potassium, sodium, iron, manganese, zinc
and copper.

35

Soil Sampling
Soil samples were collected in July 2019 and in October 2020 within each treatment
enclosure along two sampling transects and within nine 1 m by 1 m quadrat frames (Figure
2.3). Within each 1 m2 quadrat, four holes were dug to a depth of 20 cm. Soil samples were
collected at two depth intervals (0-10 cm and 10-20 cm) by excavating the sidewalls of the
holes and pooling the samples by depth for a total of 36 soil cores per enclosure and 108 soil
cores annually.

Figure 2.3. Soil sampling grid within each treatment enclosure depicting main soil sampling
points and the bulk density sampling points.
Bulk Density
In October 2020, five bulk density samples were collected within each enclosure using the
excavation method (Figure 2.3), as described by Krzic et al. 2010. The excavation method is
recommended for gravelly soil and involves digging a hole approximately 1000 cm3 in size,
lining the hole with thin plastic, then filling the hole with a measured volume of water
(Krzic et al. 2010). Bulk density samples were dried at 105°C for 24 hours and weighed.
Coarse fragments (> 2mm in diameter) were removed and bulk density was calculated as “the
mass of dry, coarse fragment-free soil per unit volume of the excavated soil” (Krzic et al.
2010).

36

Soil pH
Soil samples were oven dried at 30°C until a constant weight was reached, then sifted
through a 2 mm mesh sieve and stored in paper bags. Soil pH was measured though a 1:2.5
ratio of 2 mm sieved, dried soil and distilled water (10 grams of dried soil with 25 mL of
distilled water). The sample was shaken for 1 minute, centrifuged at 4000 rpm for 3-4
minutes, left to stand for at least 30 minutes, then the pH of the suspended water was read
with a three-point calibrated pH meter (Nag 2014).
Soil Organic Matter (SOM)
Loss on ignition (LOI) techniques were used to measure soil organic matter (SOM) content
(Wang et al. 2011). For each dried and 2 mm-sieved soil sample, approximately 1.5 g of soil
was placed into a 105°C oven for 12-16 hours to remove excess moisture, then weighed.
Soils dried at 105°C were placed into a muffle oven at 500°C for five hours, then transferred
into a desiccator for one hour before re-weighing. SOM was then calculated as the soil
weight loss between at 500°C and 105°C (Wang et al. 2011; Richardson 2015).
Soil Particulate Fractionation
Within each treatment enclosure five of the nine soil sampling points (at two sampling
depths) were randomly selected for a modified soil particulate fractionation procedure
described as ‘Par + Den 5’ by Poeplau et al. (2018) (Figure 2.4). This procedure minimizes
the disruption of POM through ‘gentle’ agitation using glass beads instead of a centrifuge.
The modifications for the Par + Den 5 method are described in the protocol from the
Summerland Research and Development Centre (SuRDC), a research centre associated with
Agriculture and Agri-food Canada (Appendix 2).
Each sample of 50 g dried, 2 mm-sieved soil was shaken overnight (16 hr) at 225 rpm on an
orbital shaker with glass beads to disperse aggregates. The suspended soils were passed
through a 63 µm sieve, then separated into 1) fine and coarse sands (> 63 µm), 2) silt and
clay particles (< 63 µm), and particulate organic matter or POM (> 63 µm). The majority of
POM in the soil was located in the fine and coarse sand fractions, therefore, the POM was
37

isolated from the fine and coarse sand fractions using an NaI (sodium iodide) solution with a
specific gravity of 1.7 g cm-3. Floating POM was then removed with a scoopula and
disposable pipette. To remove residual NaI solution, the POM was rinsed with a minimum of
200 mL of deionized water on a vacuum filtration apparatus set with a No. 4 Whatman paper.
No floating POM < 63 µm was observed in soil samples.

Figure 2.4. Modified soil fraction procedure based on the Par + Den5 method described in
Poeplau et al. 2018.
Once soil mineral fractions and POM were separated, samples were oven-dried at 65°C in
pre-weighed 1000 mL beakers until a constant weight was reached. Dried samples were
finely ground using a mortar and pestle, then a subsample of each fraction (10-15 mg) was
processed through an automated elemental analyzer (Thermo Scientific™ FlashSmart™) to
determine the % carbon and nitrogen and C-N ratios. For ease of analysis, the silt and clay
particles (< 63µm) were designated as the MAOM (Midwood et al. 2021) to compare against
the POM fraction. Therefore, subsequent data analysis does not include the fine and coarse
sand fraction. Please note that the isolation of the POM was not perfect, and there may have
been some residual POM within the fine and coarse sand fractions.

38

Statistical Analysis
The statistical software R (R Core Team 2020) was used for data analysis, and the R package
ggplot2 (Wickman 2016) was deployed to visualize data and produce figures. The alpha
value for significance was set at α = .05.
Plant community structure, plant species diversity, biomass, and soil responses were
compared between grazing treatments and the baseline (pre-treatment) data using the R
package “lmer4” (Bates et al. 2015) for linear mixed effect models (LMMs). Each response
variable was either square-root or log+1 transformed to meet the assumption of normality in
the residuals, and this assumption was checked for each model by plotting the residuals and
visually assessing their distribution. The LMMs used ‘treatment’ as the fixed effect and
‘year’ as the random effect. Given that each treatment had exactly one enclosure, spatial
dependence between points within the same enclosure was captured in part by the fixed
effect of treatment; including treatment (i.e. enclosure) in both fixed and random effects
prevented model convergence, and thus the spatial relationship was left to the fixed
components only. Furthermore, because the quadrats were randomly placed, there was no
replication for individual sample locations. This reduced the spatial dependence between
years, allowing time to be captured as the random effect of year, and space to be captured in
the fixed effect of treatment. The R package “emmeans” (Lenth et al. 2022) was used to
perform Tukey HSD post-hoc tests for pairwise comparisons between all groups. This was
only applied for response variables where LMMs detected significant contrasts between the
grazing treatment enclosures and baseline data.

39

Results
Plant Community Structure
All three types of grazing management appeared to influence the plant community at Drum
Lake pasture (Table 2.3, Figure 2.5). Total grass cover increased within all treatment
enclosures compared to baseline conditions, and doubled within the MiG enclosure from
19.1% at baseline to 38.7% after two years of grazing (Figure 2.5). The MiG and extensive
grazing enclosures differed significantly from baseline native grass cover, where both values
quadrupled (Table 2.3, Figure 2.5). The targeted enclosure also experienced a threefold
increase in native grass cover (Table 2.3, Figure 2.5). Introduced grass cover did not differ
between treatments, however extensive grazing demonstrated the lowest introduced grass
cover at 0.6%, and MiG the greatest at 5.5%.
Total and native forb cover were greatest within the targeted grazing enclosure at 63.1% and
44.9%, respectively, whereas no differences were detected between the baseline, MiG, and
extensive grazing treatments (Table 2.3, Figure 2.5). Introduced forb cover was greatest within
the targeted enclosure at 18.3%. No significant differences in spotted knapweed cover were
observed.
Soil crust and litter cover did not differ between grazing treatments; however, litter cover was
the lowest within the targeted enclosure at 14%. Furthermore, bare ground cover decreased
significantly within the targeted grazing enclosure when compared to the extensive and MiG
enclosure (Table 2.3, Figure 2.5).

40

Table 2.3. Summarized linear mixed effect models for the vegetation composition and ground
cover response variables, including the transformed estimated means and standard error
values. Significant contrasts (α = .05) between treatments are provided from Tukey HSD
post-hoc tests.
Estimated Mean
(transformed)

Response variable

Transformation

Treatment

Grass Cover (%)
Total

square root

Control
MiG
TG
EG
Control
MiG
TG
EG
Control
MiG
TG
EG

3.8
6.14
4.89
5.69
2.1
5.63
4.51
5.6
1.12
0.98
0.82
0.19

0.671

Control
MiG
TG
EG
Control
MiG
TG
EG
Control
MiG
TG
EG
Control
MiG
TG
EG

5.43
4.88
7.76
6.2
4.21
3.19
6.33
4.8
3.1
3.28
3.6
3.34
1.97
1.95
2.07
1.97

0.331

Control
MiG
TG
EG
Control
MiG
TG
EG
Control
MiG
TG
EG

2.42
2.27
1.55
2.27
2.09
2.52
1.67
2.74
3.96
4.21
3.68
3.85

0.464

Native

Introduced

Forb Cover (%)
Total

Native

Introduced

Spotted knapweed

Ground Parameters (%)
Bare ground

Crust

Litter

square root

log+1

square root

square root

square root

log+1

log+1

square root

square root

Standard error

Significant contrast

MiG
Control

0.553
0.783
0.625
0.415
0.368

0.405
1.003
0.812
0.293
0.35
0.221
0.243

0.352
0.347
0.343
0.197
0.242

MiG; EG
Control
Control
TG
TG
MiG; EG; Control
TG
TG
MiG
TG
MiG; EG
TG
-

41

A

B

C

D

C

E

Figure 2.5. Violin plots illustrating the untransformed (A) total grass, (B) native grass, (C)
total forb, (D) native forb, and (E) bare ground percent cover data. Comparisons are made
between the pre-treatment (control) data and the three grazing treatments after two years of
treatments. Different letters between treatment denotes significance (α = .05) from Tukey
HSD post-hoc tests.

42

Diversity Indices
Shannon (H’) and Simpson (D) diversity increased amongst all grazing treatments when
compared to the baseline data (Table 2.4). H’ and D diversity both increased significantly
within the MiG and extensive grazing enclosures, whereas the targeted enclosure experienced
a significant increase in H’ but did not differ from baseline for D diversity (Table 2.4, Figure
2.6).
Table 2.4. Summarized linear mixed effect models for the Shannon (H’) and Simpson (D)
diversity index response variables, including the transformed estimated means and standard
error values. Significant contrasts (α = .05) between treatments are provided from Tukey
HSD post-hoc tests.
Response variable

Transformation

Treatment

Diversity Index
Shannon (H')

log+1

Control
MiG
TG
EG
Control
MiG
TG
EG

Simpson (D)

log+1

Estimated Mean
(transformed)
0.803
0.961
0.949
0.979
0.522
0.584
0.562
0.584

A

Standard error

0.022
0.027
0.011
0.013

Significant contrast

MiG; TG; EG
Control
Control
Control
MiG; EG
Control
Control

B

Figure 2.6. Violin plots illustrating the untransformed (A) Shannon (H) diversity and, (B)
Simpson (D) diversity. Comparisons are made between the pre-treatment (control) data and
the three grazing treatments after two years of treatment. Different letters between treatment
denotes significance (α = .05) from Tukey HSD post-hoc tests.

43

Biomass
Biomass values varied by functional group and between grazing treatments. The targeted
grazing enclosure displayed the greatest total live biomass, at 54.6 g/.25m2, whereas the MiG
and extensive enclosures contained similar values of 36.8 and 36.4 g/.25m2, respectively
(Table 2.5, Figure 2.7). The extensive enclosure had the lowest grass productivity at 14.4
g/.25m2, whereas the MiG enclosure had the greatest at 21.3 g/.25m2 (Table 2.5). Mixed forb
productivity was greatest within the targeted enclosure, and the lowest within the MiG
enclosure (Table 2.5, Figure 2.7). Spotted knapweed biomass did not vary between treatment
enclosures. Litter biomass was greatest within the targeted enclosure, whereas the MiG and
extensive enclosures contained approximately half the litter of the targeted enclosure (Table
2.5, Figure 2.7).
Table 2.5. Summarized linear mixed effect model results for the biomass data, including the
transformed estimated means and standard error values. Significant contrasts (α = .05)
between treatments are provided from Tukey HSD post-hoc tests.
Response variable

Transformation

Treatment

Biomass (g/.25m^2)
Mixed grasses

log + 1

MiG
TG
EG
MiG
TG
EG
MiG
TG
EG
MiG
TG
EG
MiG
TG
EG

Mixed forbs

Spotted knapweed

Total live

Litter

log + 1

log + 1

log + 1

log + 1

Estimated Mean
(transformed)
3.03
2.73
2.6
1.33
2.82
2.36
1.5
2.28
1.58
3.52
3.94
3.55
3.02
3.63
2.98

Standard error

0.106

0.231

0.409

0.154

0.132

Significant contrast

EG
MiG
TG; EG
MiG
MiG
TG
EG; MiG
TG
TG
EG; MiG
TG

The estimated biomass utilization values were greatest within the extensive and MiG
enclosures, at 54.1 and 54.2%, respectively. The targeted enclosure experienced the lowest
percent utilization at 33.5%.

44

A

B

C

D

E

Figure 2.7.Violin plots illustrating the untransformed (A) mixed grass, (B) mixed forb, (C)
spotted knapweed (SKW), (D) total live, and (E) litter biomass expressed as g/.25m2 in late
July. Different letters between treatment denotes significance (α = .05) from Tukey HSD
post-hoc tests.

45

Forage Nutritive Values
At the point of targeted grazing, C. stoebe appeared to contain more crude protein and total
digestible nutrients (TDN) in late July than the graminoids did in late July and early
September (Table 2.6). Acid detergent fiber (ADF) and neutral detergent fiber (NDF) values
were lower in spotted knapweed compared to the grasses (Table 2.6). In respect to nutrients,
spotted knapweed contained more calcium, potassium, and iron than the grasses during the
targeted grazing trial.

46

Table 2.6. Descriptive summary of mean forage nutritive values (± stdv.) of graminoids in late July (point of targeted grazing
treatment) and early September, and spotted knapweed in late July.

Plant Group

ADF1 aNDF2

Ash

% Dry Matter
Crude
Ca Protein
K

Late July

41.5
(1.0)

67.5
(2.6)

13.1
(1.6)

0.4
(0.1)

5.0
(0.4)

0.7
(.1)

0.1
(.01)

0.1
(.02)

50.5
(1.3)

15.5
(0.8)

82.5
(34.2)

36.7
(3.4)

15.5
(1.8)

Early
September

44.3
(1.1)

69.0
(2.0)

6.7
(1.2)

0.4
(0.1)

4.2
(0.5)

0.4
(.03)

0.1
(.02)

0.1
(.02)

54.8
(.5)

2.7
(0.8)

102.3
(17.4)

44.8
(13.0)

14.2
(2.3)

Late July

30.3
(2.6)

39.6
(4.2)

9.6
(2.5)

1.3
(0.3)

7.2
(0.8)

1.7
(.26)

0.2
(.02)

0.2
(.02)

62.2
(3.1)

8.5
(1.0)

117.7
(41.5)

38.5
(9.5)

17
(1.4)

Sampling
Period

Ppm
Mg

P

TDN3

Cu

Fe

Mg

Zn

Graminoids

Spotted
Knapweed
1

Acid detergent fiber.

2

Neutral detergent fiber.

3

Total digestible nutrients.

Soil Bulk Density
After three years of grazing, no significant differences in bulk density between treatment
groups were detected.
Soil pH
Soil pH was more acidic within the targeted enclosure compared to the MiG enclosure at
both sampling depths (Figure 2.8), whereas pH values did not differ between MiG and
extensive treatments in the 0-10 cm nor 10-20 cm sampling depths (Table 2.7, Figure 2.8).
Table 2.7. Summarized linear mixed effect models for the soil pH and soil organic matter
data at both sampling depths (0-10 and 10-20 cm), including the transformed estimated
means and standard error values. Significant contrasts (α = .05) between treatments are
provided from Tukey HSD post-hoc tests.
Estimated Mean
(transformed)

Standard error

MiG
TG
EG
MiG
TG
EG

2.07
2.04
2.07
2.07
2.05
2.07

0.0195
0.0195
0.0195
0.0129
0.0129
0.0129

TG
MiG; EG
TG
TG
MiG
-

MiG
TG
EG
MiG
TG
EG

1.87
2.17
2
1.67
1.71
1.57

0.148
0.148
0.148
0.141
0.141
0.141

TG
EG; MiG
TG
EG
TG

Response variable

Transformation

Treatment

pH
0-10 cm

log + 1

10-20 cm

log + 1

Soil Organic Matter
(SOM) %
0-10 cm

log + 1

10-20 cm

log + 1

A

Significant contrast

B

Figure 2.8.Violin plots illustrating soil pH at the (A) 0-10 cm and (B) 10-20 cm sampling
depths. Different letters between treatment denotes significance (α = .05) from Tukey HSD
post-hoc tests.
48

Soil Organic Matter (SOM)
Soil organic matter values ranged from 3.97% to 8.01 % within the top 20 cm of the soil
profile (Table 2.7, Figure 2.9). The majority of SOM was measured within the upper 10 cm
of soil (Table 2.7, Figure 2.9). At the 0-10 cm sampling depth, the MiG and extensive
treatment enclosures contained less SOM than the targeted treatment (Table 2.7, Figure. 2.9).
At the 10-20 cm sampling interval, the targeted enclosure contained more SOM than the
extensive enclosure, but did not differ from the MiG enclosure (Table 2.7, Figure 2.9).

A

B

Figure 2.9. Violin plots illustrating the percent soil organic matter data at the (A) 0-10 cm
and (B) 10-20 cm sampling depths. Different letters between treatment denotes significance
(α = .05) from Tukey HSD post-hoc tests.

49

Soil Fractionation
No differences in POM carbon (C), nitrogen (N) or C-N ratio were detected between
treatments at both soil sampling depths (Table 2.9). The majority of soil carbon was observed
within the POM fraction and POM-C content did not appear to change with depth (Table
2.8). POM-N did not differ by grazing treatment, but did appear to decrease with depth
(Table 2.8). The mean C-N ratio for the POM fraction varied between 14.1 to 17.3 within the
top 20 cm of soil (Table 2.8).

The C and N within the MAOM fraction was greater within the targeted grazing enclosure at
the 0 – 10 cm sampling depth (Table 2.8, Figure 2.10). No differences in soil C, N or C-N
ratio were detected in the MAOM fraction at the 10 – 20 cm sampling depth (Table 2.8).

50

Table 2.8. Summarized linear mixed effect model the soil carbon, nitrogen and C-N ratio for
the particulate organic matter (POM) and mineral associated organic matter (MAOM) by
treatment and soil sampling depth (0-10 or 10-20 cm). Significant contrasts (α = .05)
between treatments are provided from Tukey HSD post-hoc tests.
Estimated Mean
(transformed)

Standard error

MiG
TG
EG
MiG
TG
EG
MiG
TG
EG

4.52
4.73
4.59
1.2
1.26
1.22
3.78
3.75
3.77

0.109
0.109
0.109
0.045
0.045
0.045
0.064
0.064
0.064

-

MiG
TG
EG
MiG
TG
EG
MiG
TG
EG

4.52
4.6
4.47
1.09
1.14
1.13
4.16
4.05
3.95

0.103
0.103
0.103
0.022
0.022
0.022
0.066
0.066
0.066

-

MiG
TG
EG
MiG
TG
EG
MiG
TG
EG

1.69
1.89
1.7
0.553
0.623
0.554
3.08
3.07
3.1

0.064
0.064
0.064
0.069
0.069
0.069
0.266
0.266
0.266

TG
EG; MiG
TG
TG
EG; MiG
TG
-

MiG
TG
EG
MiG
TG
EG
MiG
TG
EG

1.49
1.48
1.4
0.46
0.48
0.446
3.25
3.1
3.15

0.048
0.048
0.048
0.034
0.034
0.034
0.156
0.156
0.156

-

Response variable

Transformation

Treatment

POM (0-10 cm)
Carbon

Square root

Nitrogen

Square root

C-N Ratio

Square root

POM (10-20 cm)
Carbon

Square root

Nitrogen

Square root

C-N Ratio

Square root

MAOM (0-10 cm)
Carbon

Square root

Nitrogen

Square root

C-N Ratio

Square root

MAOM (10-20 cm)
Carbon

Square root

Nitrogen

Square root

C-N Ratio

Square root

Significant contrast

51

A

B

Figure 2.10. Violin plots illustrating the (A) % carbon and (B) % nitrogen content within the
mineral associated organic matter (MAOM) at the 0-10 cm sampling depth interval.
Different letters between treatment denotes significance (α = .05) from Tukey HSD post-hoc
tests.

52

DISCUSSION
The results provide partial support for the hypothesis that intensive grazing management
practices would improve the condition of invaded, semi-arid rangelands.
Plant Community Composition and Diversity
The first prediction was that intensively-grazed pastures would demonstrate increased cover
of native plant species and litter, increased productivity, reduced bare ground cover, and
improved diversity indices when compared to extensive grazing. I also predicted that
extensive grazing would exhibit greater introduced plant species cover.
Contrary to my first hypothesis, MiG and extensive treatments both experienced significant
increases in native grass cover and improved diversity indices when compared to the baseline
(pre-treatment) conditions. Increased native grass cover suggests an improvement in range
condition, where their expansive root structures also help stabilize soil and increase organic
matter (Burke et al. 1998; Lund 2007; Li et al. 2017; Teague et al. 2020). Furthermore,
Shannon (H’) and Simpson (D) diversity both increased within all grazing treatment
enclosures when compared to baseline. I anticipated to see lower diversity within the
extensive grazing enclosure because of the potential for repeated grazing of preferred forage
plants and increased bare soil, however, the opposite was true. These results could be
attributed to the year-long rest period in between grazing events or the timing of grazing.
It is well established that incorporating rest into rangeland management plans is critical for
plant recovery following a disturbance or defoliation event, such as grazing (Sanderman et al.
2015; Augustine et al. 2020). A case study conducted in Colorado’s shortgrass prairie found
that postponing grazing for an entire growing season promoted the growth of key forage
species, such as needle and thread grass (Hesperostipa comata) and western wheatgrass
(Pascopyrum smithii) when compared to a deferment period of several weeks to months
(Derner et al. 2022). Moreover, the timing of grazing may have impacted plant community
responses to grazing. In this study, the MiG and extensive treatments were conducted in midSeptember; at this point, all preferred perennial bunchgrasses were dormant, had set seeds,
and were not as vulnerable to defoliation when compared to the active growing period in the
53

spring (Meays et al. 2014). Species, such as bluebunch wheatgrass (Pseudoroegneria
spicata) and Junegrass (Koeleria macrantha), produce regrowth when summer temperatures
drop and moisture increases in the fall (Wikeem et al. 1993). This is particularly relevant to
my study in September 2019, where 59 mm of rainfall was observed. Therefore, these
species may be able to exhibit valuable regrowth and recover more readily following a
grazing event in mid-September.
With respect to the improved diversity indices amongst all treatments, Morris (2021)
conducted a meta-analysis on the benefits and shortcomings of short-duration, moderate to
high-intensity grazing and discovered that plant species richness generally increased
compared to ungrazed areas. Moreover, the diversity and abundance of perennial grasses
increased, more so than perennial forbs in response to forms of management-intensive
grazing (Huruba et al. 2018; Morris 2021). That being said, overall species diversity and
richness was generally unaffected within semi-arid rangeland environments (Olivia et al.
2021; Morris 2021). The improved diversity indices observed in this study could therefore be
explained by the year-long grazing deferment, rather than the grazing treatment itself.
Species richness, or the number of unique species within an ecosystem, is also related to
plant productivity (Grime 1973; Graham and Duda 2011; Fraser et al. 2015). The humpbacked model (Grime 1973) suggests that there is a unimodal relationship between plant
productivity and species richness. Ecosystems that are drought-prone and contain fewer
nutrients may experience lower plant productivity and a limited number of species adapted to
those conditions (Grime 1973; Graham and Duda 2011; Fraser et al. 2015). Whereas the
number of species within productive ecosystems may be limited by competitive effects
(Graham and Duda 2011; Fraser et al. 2015). In this experiment, a combination of grazing
and grazing deferment may have stimulated increased plant productivity, and, according to
Grime’s hump-backed model (1973), should therefore correspond with increased species
richness and diversity.
The introduced grass cover within all treatment enclosures appeared to decrease when
compared to baseline conditions, which is encouraging in respect to maintaining productive
and preferred forage species. Introduced grasses, such Kentucky bluegrass (Poa pratensis),

54

are a concern in rangelands because they have features that allow them to tolerate heavy
grazing pressure, such as low growing points and the production of underground rhizomes
(Meays et al. 2010). Cheatgrass (Bromus tectorum) also has an advantage in that it is an
annual, it can germinate in the spring, summer, or fall, and it develops large, sharp awns
when mature making it unpalatable for grazing animals (Mack 1981; Larson et al. 2017). The
presence of these species highlights the importance of monitoring and adaptation when
implementing a new grazing management strategy, and ensuring the percent cover of
unwanted species does not increase over time.
The targeted grazing enclosure demonstrated the greatest total forb cover; however, the cover
of native forbs did not significantly differ from baseline conditions. Two of the dominant
native forbs on site, common yarrow (Achillea millefolium) and silky lupine (Lupinus
sericeus) are characterized as ‘increaser’ species (Vesk and Westoby 2001; Gayton 2003).
The term ‘increaser’ and ‘decreaser’ correspond to a plant’s relative abundance or growth in
response to increased grazing pressure (Vesk and Westoby 2001). Interestingly, I observed
many grazed lupine and yarrow plants within the targeted enclosure (Figure 2.11). Lupines
are known to be toxic to cattle when ingested in large quantities (Lee et al. 2020) and cattle
generally tend to avoid grazing them (Lloyd 2019); therefore, I was surprised at the amount
of native forb biomass removed on-site. It appeared that the cattle either grazed the native

A

B

Figure 2.11. A landscape view of the targeted grazing treatment enclosure before (A) and
after (B) grazing at the late bud to flowering stage in early August, 2020.
and non-native forbs selectively over the dormant grasses at the point of targeted grazing, or
grazed the forbs because that was the most available type of forage within the targeted
enclosure. Either way, the increased grazing pressure placed on the dominant forb species at
the point of targeted grazing may have stimulated their growth and spread.
55

Neither litter nor biological crust cover differed between grazing treatments. However, bare
ground cover was reduced within all three grazing treatments compared to baseline
conditions. Bare ground is exposed soil that is unprotected by plant matter or organic
materials, and can experience increased temperatures, decreased microbial activity, and
increased erosion risk (Teague et al. 2011; Porensky et al. 2020). Research conducted by
Teague et al. (2011) found that heavy, continuous season-long grazing increased bare
ground, which can subsequently lead to nutrient leaching and increased opportunities for the
establishment of invasive plant species, of which are often adapted to low nutrient and
moisture conditions (Lee et al. 2014; Steanowicz et al. 2018; Fenetahun et al. 2021).
Therefore, the reduction in bare ground observed in this study may help maintain or decrease
soil temperatures, improve soil microbial activity, limit the amount of soil water being
evaporated, and reduce the risk of plant invasion (Jefferson and Maxwell 2015; Porensky et
al. 2020). I anticipated increased bare ground cover within the extensive grazing enclosure,
however, the observation in this experiment may be associated with the prolonged period of
rest and opportunity for plant establishment during active spring growth. If I had timed the
grazing in the spring, I may have observed a different result.
Biomass
Grazing intensity refers to the amount of biomass or forage that has been removed during the
grazing period (Wikeem and Wikeem 2004; Gardner et al. 2013). The estimated biomass
utilization rates for the MiG and extensive enclosures were very similar at 54.1% and 54.2 %,
respectively. The utilization rate exceeded the ‘take half, leave half’ rule of thumb for
management-intensive grazing, which recommends 50% or more of aboveground biomass
should remain post-grazing, otherwise above and below ground growth can become
suppressed over time (Gerrish 2005). Interestingly, despite having the same stocking rate as
the MiG treatment, the targeted enclosure exhibited a lower utilization rate of 33%. The
targeted enclosure demonstrated a greater amount of live biomass; therefore, the cattle could
have consumed equal amounts of biomass within the MiG and targeted enclosures, but the
percent difference was not as great because more biomass existed in the targeted enclosure.

56

The total live, litter, and spotted knapweed biomass did not vary between the extensive and
MiG treatments. However, the MiG enclosure demonstrated the greatest grass and the lowest
forb biomass, whereas the opposite result was observed within the extensive enclosure. The
cattle within the extensive enclosure had more time to revisit and re-graze preferred plants
(Briske et al. 2008; Metera et al. 2010), that is the mixed grasses, at the point of grazing. The
cattle within the MiG enclosure may have grazed more uniformly due to the short grazing
duration and greater animal numbers (Conant et al. 2003). Stubble height measurements on
the mixed grass and forb communities at the point of the MiG and extensive treatments
would have provided a clearer description the grass versus forb use.
Total litter biomass was greatest within the targeted grazing enclosure. This result could be
explained by the way litter biomass was collected, or because of the greater amount of total
live biomass within the targeted enclosure. I combined the standing and ground litter
together; therefore, dead-standing, woody knapweed stalks may have contributed more litter
weight within the targeted enclosure.
It is important to note that biomass utilization, or the amount of forage eaten or trampled by
grazing animals, is a difficult variable to accurately measure (Pauler et al. 2020; Jansen et al.
2021). Natural, semi-arid grasslands are among the most diverse grassland types in the world,
and that diversity can be represented at very small spatial scales (Pauler et al. 2020).
Variation in biomass productivity is, therefore, expected within these ecosystems. This
variation, combined with our moderately-sized research pastures (50 m x 50 m) suggest that
there may have been patches within each treatment enclosure that cattle selected more
heavily than others. Consequently, the estimated utilization rates calculated in this chapter
should be interpreted with caution.
Forage Nutritive Values
Spotted knapweed was originally believed to be an unpalatable species to grazing livestock
(Griffith and Lacey 1991; Sheley and Jacobs 1997; Sheley et al. 1998). However, the results
of this study demonstrate that, when grazed at the appropriate time, knapweed can be a
valuable source of forage. This is due to the high crude protein and relatively low fibre
values, findings which indicate good quality forage (Zhai et al. 2018). The results of this
57

forage nutritive analysis are coincident findings from Kelsey and Mihalovich (1987), Olson
et al. (1997), and Ganguli et al. (2010).
Targeted Cattle Grazing and Spotted Knapweed
The second hypothesis was that targeted cattle grazing would reduce the cover and
productivity of the invasive plant, spotted knapweed (Centaurea stoebe). I found that the
cover and productivity of spotted knapweed increased after two targeted grazing applications
in late-July. Spotted knapweed cover nearly doubled compared to the baseline plant
community, and productivity was also greatest within the targeted enclosure at 14.5g/.25m2.
It is therefore important to consider the compensatory growth effect, where plants will grow
or increase in productivity in response to a disturbance or defoliation event, such as grazing
(Kimball and Schiffman 2003). Compensatory growth is a common trait in invasive plant
species, and is likely triggered by defoliation of the apical meristem or other disturbances
(Müller-Schärer et al. 2004). Moreover, the compensatory growth of reproductive spotted
knapweed stems has been reported after targeted sheep grazing (Olson et al. 1997), but
overall grazing can still reduce knapweed dominance over time (Henderson et al. 2012;
Mosley et al. 2016; Marchetto et al. 2021).

If my study was conducted over a longer time period, perhaps I would have seen a reduction
in spotted knapweed cover and productivity. For instance, Mosley et al. (2016) observed an
86% reduction in spotted knapweed density over a four-year period using targeted sheep
grazing in late-July. Furthermore, if my study was repeated with additional knapweed control
measures, such as a fall herbicide application to target residual rosettes and ungrazed mature
plants, we may also have seen a reduction plant density and productivity.

The forage nutrition analysis on spotted knapweed and the grasses at the point of targeted
grazing may explain the selectivity of spotted knapweed observed (Figure 2.11). At this time,
spotted knapweed is still actively growing, green and leafy, and therefore more palatable than
the dormant perennial bunchgrasses. In the field we also observed many bunchgrasses that
were either ungrazed or grazed minimally with sufficient residual biomass (Figure. 2.11).

58

These findings are similar to Henderson et al. (2018), where sheep appeared to prefer and
select for spotted knapweed and avoid the dormant grasses.

In respect to a targeted grazing strategy, monitoring grazing intensity and adverse effects on
the native plant community is very important. Fortunately, I did not observe significant
reductions in native grass or native forb cover after two years of targeted grazing. However,
to ensure there are no adverse effects on the native vegetation, future research should involve
longer study periods (> 5) with annual monitoring before and after targeted grazing
treatments. Other research opportunities include: mixed-species targeted grazing and
integrating multiple rangeland restoration activities, such as seeding or re-vegetation postgrazing. Recommendations for spotted knapweed treatment followed by targeted grazing is
discussed in Chapter 4, “Research Summary and Management Implications”.
Soil Responses to Grazing Management
Finally, I predicted that extensive grazing would increase soil bulk density and decrease soil
organic matter, carbon, and nitrogen. Contrary to the hypothesis, both of the fall-grazed
treatments, management-intensive and extensive grazing, appeared to exhibit similar
responses to grazing when compared to targeted grazing.
No differences in soil bulk density were detected amongst the grazing treatments in this
study, however, bulk density can increase in response to moderate to heavy grazing pressure
(Chanasyk and Naeth 1995; Lee et al. 2014; Lai and Kumar 2020). A study conducted in
Alberta’s foothill fescue grasslands found that an annual short-duration (1-week), heavy
grazing period in mid-June increased soil bulk density by approximately 9% over four years
(Chanasyk and Naeth 1995). In the fall, there is a tendency for greater soil moisture and
precipitation within B.C.’s Southern Interior compared to mid-summer. Increased soil
moisture and finer-textured soils may be attributed to increased bulk density during grazing
periods (Laycock and Conrad 1967; Mayel et al. 2020). Finer-textured soil particles also
display greater surface tension and can hold onto water more tightly, therefore soil is more
easily compressed whereas dry soil requires much more force to become compacted
(Chanasyk and Naeth 1995; Mayel et al. 2020). It is important to note that just five bulk
59

density samples were collected within each treatment enclosure in this study. A greater
sample size would have provided a more representative sample to compare against, and
baseline samples would have allowed me to detect treatment effects more clearly.
Moreover, soil pH was more acidic within the targeted enclosure at both soil sampling depths
(0-10 and 10-20 cm) when compared to the fall-grazed treatments. Soil pH can be influenced
in several ways such as nitrogen additions (Lavallee et al. 2020) or the disruption or removal
of topsoil in which the subsoil could be more alkaline (Evans et al. 2014). A global metaanalysis conducted by Lai and Kumar (2020) examined the influence grazing on soil pH and
found that livestock manure and urine can generally increase soil pH. Without baseline soil
data, however, it is difficult to interpret whether the decrease in soil pH was a result of the
targeted grazing treatment, or whether the soil and composition of the parent material within
the targeted enclosure is less alkaline than the other pastures.
Changes in soil organic matter, carbon, and nitrogen are likely associated with the changes in
net primary productivity or biomass production (Conant et al. 2003). The results in this study
demonstrate that the majority of SOM is located within the upper 10 cm of soil and that it
decreases with depth. Plant productivity and SOM content was greatest within the targeted
grazing enclosure. Moreover, SOM generally increases in response to the quantity and
decomposition rate of plant litter (Ghorbani and Raiesi 2012; Teague et al. 2013; Lavallee et
al. 2020). No significant difference in soil carbon, nitrogen or C-N ratio by grazing treatment
were detected in this study, although POM was found to be the dominant form of carbon
within the semi-arid rangeland soils. Midwood et al. (2021) detected similar results with
natural grassland soils collected in the Okanagan region of B.C, where POM contributed
69% of the total carbon. This result could be attributed to climatic regime of the interior,
semi-arid rangelands (Midwood et al. 2021). Furthermore, the C-N ratios within the MAOM
fraction were much less than that of the POM fraction, which corresponds with findings from
Cotrufo et al. (2019).
Mineral weathering processes are relatively slow within semi-arid environments when
compared to irrigated agricultural pastures or pastures with higher soil moisture (Byrnes et al.
2018; Crème et al. 2020); therefore, the decomposition and storage of soil organic matter and

60

soil carbon may become delayed. Interestingly, the MAOM C and N appeared to be greater
within the targeted enclosure at the 0-10 cm sampling depth. MAOM is considered to be the
more stable and protected form of carbon, where the turnover time or rate of microbial
decomposition and process of sorption to the mineral-sized particles is very complicated and
slow (Lavallee et al. 2020; Midwood et al. 2021). Additionally, no baseline soil data was
collected, so it may be that the soil within the targeted enclosure simply exhibited different
chemical properties.
Conclusions & Future Research
Overall, this study demonstrated that grazing can strongly influence the composition and
productivity of plants in semi-arid environments. It is well documented that repeated grazing
on preferred forages such as native bunchgrasses, can reduce plant vigor (Briske et al. 2008;
Metera et al. 2010; Pauler et al. 2020). Therefore, it is also important to consider animal
behaviour in this study. Cattle tend to be selective when grazing; their choices depend on
forage availability, species distribution, and whether there are containment measures, such as
fencing, in place (Pauler et al. 2020). The cattle used in this study were previously exposed to
electric fencing, however, it is uncertain whether they had experienced being in small animal
numbers, such as in the extensive grazing period. Moreover, changes in environmental or
social conditions may influence animal behaviour and performance, specifically feed intake
(Fraser 2004; Pauler et al. 2020). Efforts were made to minimize animal stress (providing
water, ensuring sufficient forage was available, paired cattle/no isolated individuals),
however it is unclear whether feeding behaviours would be altered within the 50 m x 50 m
enclosures.
Furthermore, a mesic rangeland site may have experienced more pronounced vegetation
responses to the grazing treatments used in this study (Teague et al. 2008; Booker et al.
2013). Greater moisture availability often equates to a more productive rangeland, whereas
xeric sites, or those that experience minimal rainfall and sporadic growing seasons, generally
have lower productivity (Teague et al. 2008). Therefore, Teague et al. (2008) posits that xeric
or semi-arid pastures that are allowed longer periods of grazing deferment and plant recovery
may not experience as pronounced benefits when grazing intensively when compared to a

61

site with greater moisture availability. This experiment should therefore be repeated on a site
with greater soil moisture.
In regard to the soil sampling, digging a soil pit within each enclosure and analyzing the
profile would have been helpful to better understand the soil properties. This analysis also
may have better explained the differences I observed in pH and MAOM C content. The
short-term impacts of grazing management on belowground soil processes, such as microbial
activity and decomposition rates, are also not very well understood within semi-arid
rangelands (Whitehead 2020; Shrestha et al. 2020). Exploring microbial community
assemblages and measuring the degree of weathering in these areas are important in our
understanding of long-term soil carbon sequestration. Another variable that was not
measured in this study was soil moisture. Soil moisture can influence the rate of organic
matter decomposition, and is an important variable to consider when planning future soil
carbon sequestration studies.
This study lacked replication of grazing treatments; therefore, this should be considered
when interpreting results. However, with a limited timeline and only so many research
assistants available, it was not feasible to conduct a replicate set of grazing pastures.
Furthermore, baseline data, or data that was collected prior to the grazing manipulations, was
not collected for soil properties or biomass data. It is therefore unclear whether the observed
changes were a result of the grazing treatment or if differences in biomass or soil
characteristics existed prior to the grazing manipulations (Bull et al. 2014). Despite these
limitations, I was able to provide a thorough investigation of site characteristics and did the
best I could with the resources I had.
In conclusion, it is important to consider that management-intensive grazing is complex and
can require a great amount of labour and resources. For MiG to be successful, it is critical to
adapt management practices in response to weather patterns, forage availability, and to the
landscape or topography, among other factors (Savory and Parsons 1980; Dobb 2013; Mann
and Sherren 2018; Augustine et al. 2020). Setting appropriate and achievable goals,
practicing sound decision making, implementing thorough monitoring, and evaluating the

62

success of management plans is key is MiG is to be successful within semi-arid rangeland
environments (Savory and Parsons 1980; Mann and Sherren 2018; Derner et al. 2022).

63

REFERENCES
Abdalla M, Hastings A, Chadwick DR, Jones DL, Evans CD, Jones MB, Rees RM, Smith P.
2018. Critical review of the impacts of grazing intensity on soil organic carbon storage and
other soil quality indicators in extensively managed grasslands. Agric Ecosyst Environ.
253(October 2017):62–81.
Alemu A, Kröbel R, McConkey B, Iwaasa A. 2019. Effect of increasing species diversity and
grazing management on pasture productivity, animal performance, and soil carbon
sequestration of re-established pasture in Canadian Prairie. Animals. 9(4):127.
Almeida RF, Queiroz IDS, Mikhael JER, Oliveira RC, Borges EN. 2019. Enriched animal
manure as a source of phosphorus in sustainable agriculture. Int J Recycl Org Waste Agric.
8(s1):203–210.
Augustine DJ, Derner JD, Fernández-Giménez ME, Porensky LM, Wilmer H, Briske DD.
2020. Adaptive, multipaddock rotational grazing management: A ranch-scale assessment of
effects on vegetation and livestock performance in semiarid rangeland. Rangel Ecol Manag.
73(6):796–810.
Bader BR, Taban SK, Fahmi AH, Abood MA, Hamdi GJ. 2021. Potassium availability in soil
amended with organic matter and phosphorous fertiliser under water stress during maize (Zea
mays L) growth. J Saudi Soc Agric Sci. 20(6):390–394.
Bates D, Mächler M, Bolker B, Walker S (2015). “Fitting Linear Mixed-Effects Models
Using lme4.” Journal of Statistical Software, 67(1), 1–48.
Bork EW, Döbert TF, Grenke JSJ, Carlyle CN, Cahill JF, Boyce MS. 2021. Comparative
pasture management on Canadian cattle ranches with and without adaptive multipaddock
grazing. Rangel Ecol Manag. 78:5–14.
Bradfield GE, Cumming WFP, Newman RF, Krzic M. 2021. Grazing exclosures reveal
divergent patterns of change in bunchgrass grasslands of western Canada. Botany. 99(1):9–
22.
Bull JW, Gordon A, Law EA, Suttle KB, Milner-Gulland EJ. 2014. Importance of baseline
specification in evaluating conservation interventions and achieving no net loss of
biodiversity. Conserv Biol. 28(3):799–809.
Burke IC, Laurenroth WK, Vinton MA, Hook PB, Kelly RH, Epstein HE, Aguiar MR,
Robles MD, Aguilera MO, Murphy KL, et al. 1998. Plant-soil interactions in temperate
grasslands. Biogeochemistry. 42:121–143.
Byrnes RC, Eastburn DJ, Tate KW, Roche LM. 2018. A global meta-analysis of grazing
impacts on soil health indicators. J Environ Qual. 47(4):758–765.

64

Chanasyk DS, Naeth MA. 1995. Grazing impacts on bulk density and soil strength in the
foothills fescue grasslands of Alberta, Canada. Can J Soil Sci. 75(4):551–557.
Chen W, Huang D, Liu N, Zhang Y, Badgery WB, Wang X, Shen Y. 2015. Improved
grazing management may increase soil carbon sequestration in temperate steppe. Sci Rep.
5:1–13. doi:10.1038/srep10892.
Clements DR, Krannitz PG, Gillespie SM. 2007. Seed bank responses to grazing history by
invasive and native plant species in a semi-desert shrub-steppe environment. Northwest Sci.
81(1):37–49.
Coffey L. 2007. Targeted Grazing: A Natural Approach to Vegetation Management and
Landscape Enhancement. Launchbaugh KL, editor. Centennial, CO: American Sheep
Industry Association.
Conant RT, Paustian K, Elliott ET. 2001. Grassland management and conversion into
grassland: effects on soil carbon. Ecol Appl. 11(2):343–355.
Conant RT, Paustian K. 2002. Potential soil carbon sequestration in overgrazed grassland
ecosystems. Global Biogeochem Cycles. 16(4):90-1-90–9.
Conant RT, Six J, Paustian K. 2003. Land use effects on soil carbon fractions in the
southeastern United States. I. Management-intensive versus extensive grazing. Biol Fertil
Soils. 38(6):386–392.
Cotrufo MF, Ranalli MG, Haddix ML, Six J, Lugato E. 2019. Soil carbon storage informed
by particulate and mineral-associated organic matter. Nat Geosci. 12(12):989–994.
doi:10.1038/s41561-019-0484-6. http://dx.doi.org/10.1038/s41561-019-0484-6.
Crème A, Rumpel C, Malone SL, Saby NPA, Vaudour E, Decau M-L, Chabbi A. 2020.
Monitoring grassland management effects on soil organic carbon—a matter of scale.
Agronomy. 10(12):2016.
de Moraes Sá JC, Potma Gonçalves DR, Ferreira LA, Mishra U, Inagaki TM, Ferreira Furlan
FJ, Moro RS, Floriani N, Briedis C, de Oliveira Ferreira A. 2018. Soil carbon fractions and
biological activity based indices can be used to study the impact of land management and
ecological successions. Ecol Indic. 84(January):96–105.
DiTomaso JM. 2000. Invasive weeds in rangelands: Species, impacts, and management.
Weed Sci. 48(2):255–265.
Dobb, A. 2013. BC farm practices and climate change adaptation: management-intensive
grazing. BC Agriculture and Food Climate Action Initiative. 22p.
Döbert TF, Bork EW, Apfelbaum S, Carlyle CN, Chang SX, Khatri-Chhetri U, Silva
Sobrinho L, Thompson R, Boyce MS. 2021. Adaptive multi-paddock grazing improves water
infiltration in Canadian grassland soils. Geoderma. 401(February).
65

[ECCC] Environment and Climate Change Canada. [Internet]. 2021. Historical data for
Merritt, BC. Climate ID #1125073. www.climate.weather.gc.ca; [updated 2021 Nov 25].
Evans CRW, Krzic M, Broersma K, Thompson DJ. 2012. Long-term grazing effects on
grassland soil properties in southern British Columbia. Can J Soil Sci. 92(4):685–693.
Fraser LH, Carlyle CN. 2011. Is spotted knapweed (Centaurea stoebe L.) patch size related
to the effect on soil and vegetation properties? Plant Ecol. 212(6):975–983.
Ganguli AC, Hale MB, Launchbaugh KL. 2010. Seasonal change in nutrient composition of
spotted knapweed and preference by sheep. Small Rumin Res. 89(1):47–50.
Gardner, W., M. Khadka and V. Volpatti. 2013. The grazing response index project.
Agriculture and Agri-Food Canada. Kelowna, BC: Thompson Rivers University, 39 pp.
Gayton D V. 2003. British Columbia grasslands: Monitoring vegetation change. FORREX–
Forest Research Extension Partnership, Kamloops, B.C. FORREX Series 7.
Gayton D, Miller V. 2012. Impact of biological control on two knapweed species in British
Columbia. Ecosyst Manag. 13(3):1–14.
Graham JH, Duda JJ. 2011. The humpbacked species richness-curve: A contingent rule for
community ecology. Int J Ecol. 2011(Figure 1). doi:10.1155/2011/868426.
Griffith D, Lacey JR. 1991. Economic evaluation of spotted knapweed (Centaurea maculosa)
control using picloram. J Range Manag. 44(1):43. doi:10.2307/4002636.
Grime JP. 1973. Competitive exclusion in herbaceous vegetation. Nature. 242:344–347.
Gupta KK, Aneja KR, Rana D. 2016. Current status of cow dung as a bioresource for
sustainable development. Bioresour Bioprocess. 3(1).
Henderson SL, Mosley TK, Mosley JC, Kott RW. 2012. Spotted knapweed utilization by
sequential cattle and sheep grazing. Source Rangel Ecol Manag. 65(3):286–291.
Hewins DB, Lyseng MP, Schoderbek DF, Alexander M, Willms WD, Carlyle CN, Chang
SX, Bork EW. 2018. Grazing and climate effects on soil organic carbon concentration and
particle-size association in northern grasslands. Sci Rep. 8(1):1–9.
Hobbs RJ, Huenneke LF. 1992. Disturbance, diversity, and invasion: implications for
conservation. Conserv Biol. 6(3):324–337. doi:10.1046/j.1523-1739.1992.06030324.x.
Iverson K. 2004. Grasslands of the southern interior. Victoria, BC: Ministry of Sustainable
Resource Management. 6p.
Jansen V, Traynor ACE, Karl JW, Lepak N, Sprinkle J. 2021. Monitoring grazing use:
Strategies for leveraging technology and adapting to variability. Rangelands. 44(1):64–77.
66

Jefferson J, Maxwell R. 2015. Evaluation of simple to complex parameterizations of bare
ground evaporation. J Adv Model Earth Syst. 7:1075–1092.
Kimball S, Schiffman PM. 2003. Differing effects of cattle grazing on native and alien
plants. Conserv Biol. 17(6):1681–1693.
Klinkenberg, Brian. (Editor) 2021. E-Flora BC: Electronic atlas of the flora of British
Columbia [eflora.bc.ca]. Lab for Advanced Spatial Analysis, Department of Geography,
University of British Columbia, Vancouver.
Krzic M, Naugler T, Dyanatkar S, and Crowley C. 2010. Virtual soil lab modules. University
of British Columbia (UBC), Vancouver.
Krzic M, Lamagna SF, Newman RF, Bradfield G, Wallace BM. 2014. Long-term grazing
effects on rough fescue grassland soils in southern British Columbia. Can J Soil Sci.
94(3):337–345.
Lai L, Kumar S. 2020. A global meta-analysis of livestock grazing impacts on soil properties.
PLoS One. 15(8):e0236638.
Lal R. 2009. Challenges and opportunities in soil organic matter research. Eur J Soil Sci.
60(2):158–169.
Larson CD, Lehnhoff EA, Rew LJ. 2017. A warmer and drier climate in the northern
sagebrush biome does not promote cheatgrass invasion or change its response to fire.
Oecologia. 185(4):763–774.
Lavallee JM, Soong JL, Cotrufo MF. 2020. Conceptualizing soil organic matter into
particulate and mineral-associated forms to address global change in the 21st century. Glob
Chang Biol. 26(1):261–273.
Lavallee JM, Soong JL, Cotrufo MF. 2020. Conceptualizing soil organic matter into
particulate and mineral-associated forms to address global change in the 21st century. Glob
Chang Biol. 26(1):261–273. doi:10.1111/gcb.14859.
Laycock WA, Conrad PW. 1967. Effect of grazing on soil compaction as measured by bulk
density on a high elevation cattle range. J Range Manag. 20(3):136.
Lee RN, Bradfield GE, Krzic M, Newman RF, Cumming WFP. 2014. Plant community –
Soil relationships in a topographically diverse grassland in southern interior British
Columbia, Canada. Botany. 92(11):837–845.

67

Lee ST, Welch KD, Stonecipher CA, Cook D, Gardner DR, Pfister JA. 2020. Analysis of
rumen contents and ocular fluid for toxic alkaloids from goats and cows dosed larkspur
(Delphinium barbeyi), lupine (Lupinus leucophyllus), and death camas (Zigadenus
paniculatus). Toxicon. 176(November 2019):21–29.
Lenth RV, Buerkner P, Herve M, Jung M, Love J, Miguez F, Riebl H, Signmann H. 2022.
emmeans: Estimated marginal means, aka least-squares means. R package version 2022.07.1.
https://CRAN.R-project.org/package=emmeans
Li W, Xu F, Zheng S, Taube F, Bai Y. 2017. Patterns and thresholds of grazing-induced
changes in community structure and ecosystem functioning: species-level responses and the
critical role of species traits. J Appl Ecol. 54(3):963–975.
Li J, Ramirez GH, Kiani M, Quideau S, Smith E, Janzen H, Larney F, Puurveen D. 2018.
Soil organic matter dynamics in long-term temperate agroecosystems: Rotation and nutrient
addition effects. Can J Soil Sci. 98(2):232–245. doi:10.1139/cjss-2017-0127.
Lloyd D, Iverson K, Mackenzie K, Richardson H, Jeffries L. 2019. Laurie Guichon
Memorial Grasslands Interpretive Site: A Baseline Inventory. Kamloops, BC. Grassland
Conservation Council of British Columbia. 98 p.
Lund GH. 2007. Accounting for the world’s rangelands. Rangelands. 29 (1):3–10.
Mack RN. 1981. Invasion of Bromus tectorum L. into Western North America: An ecological
chronicle. Agro-Ecosystems. 7(2):145–165.
Mann C, Sherren K. 2018. Holistic Management and adaptive grazing: A trainers’ view.
Sustain. 10(6).
Maire J, Gibson-Poole S, Cowan N, Reay DS, Richards KG, Skiba U, Rees RM, Lanigan GJ.
2018. Identifying urine patches on intensively managed grassland using aerial imagery
captured from remotely piloted aircraft systems. Front Sustain Food Syst. 2(April):1–11.
Marchetto KM, Wolf TM, Larkin DJ. 2021. The effectiveness of using targeted grazing for
vegetation management: a meta‐analysis. Restor Ecol. 1-12.
Mayel S, Jarrah M, Kuka K. 2021. How does grassland management affect physical and
biochemical properties of temperate grassland soils? A review study. Grass Forage Sci.
76(2):215–244. doi:10.1111/gfs.12512.
Meays CL, Laliberte AS, Doescher PS. 2000. Defoliation response of bluebunch wheatgrass
and crested wheatgrass: Why we cannot graze these two species in the same manner.
Rangelands. 22(6):6–8.
Metera E, Sakowski T, Słoniewski K, Romanowicz B. 2010. Grazing as a tool to maintain
biodiversity of grassland - a review. Anim Sci Pap Reports. 28(4):315–334.
68

Midwood AJ, Hannam KD, Gebretsadikan T, Emde D, Jones MD. 2021. Storage of soil
carbon as particulate and mineral associated organic matter in irrigated woody perennial
crops. Geoderma. 403:115185.
Mohammadi K, Heidari G, Khalesro S, Sohrabi Y. 2011. Soil management, microorganisms
and organic matter interactions: A review. African J Biotechnol. 10(86):19840–19849.
Mosley JC, Frost RA, Roeder BL, Mosley TK, Marks G. 2016. Combined herbivory by
targeted sheep grazing and biological control insects to suppress spotted knapweed
(Centaurea stoebe). Invasive Plant Sci Manag. 9(1):22–32.
Müller-Schärer H, Schaffner U, Steinger T. 2004. Evolution in invasive plants: Implications
for biological control. Trends Ecol Evol. 19(8):417–422.
Nag, A. 2015. Analytical and instrumental techniques in agriculture, environmental and food
engineering. Delhi, India: PHI Learning. 267p.
Newman R, Krzic M, Wallace B. 2011. Linking range health assessment methodology with
science rough fescue grasslands of British Columbia. Victoria, BC.
Olson BE, Wallander RT, Lacey JR, Olson BE, Wallander RT, Lacey JR. 1997. Effects of
sheep grazing on a spotted knapweed-infested Idaho fescue community. 50(4):386–390.
Palit R, Gramig G, Dekeyser ES. 2021. Kentucky bluegrass invasion in the northern great
plains and prospective management approaches to mitigate its spread. Plants. 10(4):1–14.
Pavlů V, Hejcman M, Pavlů L, Gaisler J, Nežerková P. 2006. Effect of continuous grazing on
forage quality, quantity and animal performance. Agric Ecosyst Environ. 113(1–4):349–355.
Pauler CM, Isselstein J, Suter M, Berard J, Braunbeck T, Schneider MK. 2020. Choosy
grazers: Influence of plant traits on forage selection by three cattle breeds. Funct Ecol.
34(5):980–992.
Poeplau C, Don A, Six J, Kaiser M, Benbi D, Chenu C, Cotrufo MF, Derrien D, Gioacchini
P, Grand S, et al. 2018. Isolating organic carbon fractions with varying turnover rates in
temperate agricultural soils – A comprehensive method comparison. Soil Biol Biochem.
125(April):10–26.
Porensky LM, McGee R, Pellatz DW. 2020. Long-term grazing removal increased invasion
and reduced native plant abundance and diversity in a sagebrush grassland. Glob Ecol
Conserv. 24:e01267.
R Core Team. 2020. R: A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. Available from: https://www.Rproject.org/.

69

Raiesi F, Asadi E. 2006. Soil microbial activity and litter turnover in native grazed and
ungrazed rangelands in a semiarid ecosystem. Biol Fertil Soils. 43(1):76–82.
Reeder JD and Schuman GE. 2002. Influence of livestock grazing on C sequestration in
semi-arid mixed-grass and short-grass rangelands. Environmental Pollution 116:457- 463.
Rhodes AC, Rutledge J, DuPont B, Plowes RM, Gilbert LE. 2021. Targeted grazing of an
invasive grass improves outcomes for native plant communities and wildlife habitat. Rangel
Ecol Manag. 75(January):41–50.
Richardson H. 2015. A comparison of random forests and linear stepwise regressions to
model and map soil carbon in south-central British Columbia grasslands using normalized
difference vegetation index based models. Thompson Rivers University.
Sanderman J, Reseigh J, Wurst M, Young MA, Austin J. 2015. Impacts of rotational grazing
on soil carbon in native grass-based pastures in southern Australia. PLoS One. 10(8):1–15.
Savory A, Parsons SD. 1980. The Savory Grazing Method. Rangelands. 2(6):234–237.
Sheley RL, Jacobs JS. 1997. “Acceptable” levels of spotted knapweed (Centaurea maculosa)
control. Weed Technol. 11(2):363–368.
Sheley RL, Jacobs JS, Carpinelli MF. 1998. Distribution, biology, and management of
diffuse knapweed (Centaurea diffusa) and spotted knapweed (Centaurea maculosa). Weed
Technol. 12(2):353–362.
Shrestha BM, Bork EW, Chang SX, Carlyle CN, Ma Z, Döbert TF, Kaliaskar D, Boyce MS.
2020. Adaptive multi-paddock grazing lowers soil greenhouse gas emission potential by
altering extracellular enzyme activity. Agronomy. 10(11):1781.
Smith P, Cotrufo MF, Rumpel C, Paustian K, Kuikman PJ, Elliott JA, McDowell R, Griffiths
RI, Asakawa S, Bustamante M, et al. 2015. Biogeochemical cycles and biodiversity as key
drivers of ecosystem services provided by soils. Soil. 1(2):665–685.
Stefanowicz AM, Majewska ML, Stanek M, Nobis M, Zubek S. 2018. Differential influence
of four invasive plant species on soil physicochemical properties in a pot experiment. J Soils
Sediments. 18(4):1409–1423.
Surber LMM, Rude ME, Roeder BL, Mosley TK, Grove A V., Walker JW, Kott RW. 2011.
Percent spotted knapweed (Centaurea stoebe) in the diets of grazing sheep. Invasive Plant
Sci Manag. 4(1):95–101.
Teague WR, Dowhower SL, Baker SA, Haile N, DeLaune PB, Conover DM. 2011. Grazing
management impacts on vegetation, soil biota and soil chemical, physical and hydrological
properties in tall grass prairie. Agric Ecosyst Environ. 141(3–4):310–322.

70

Teague R, Provenza F, Kreuter U, Steffens T, Barnes M. 2013. Multi-paddock grazing on
rangelands: Why the perceptual dichotomy between research results and rancher experience?
J Environ Manage. 128:699–717.
Teague R, Kreuter U. 2020. Managing Grazing to Restore Soil Health, Ecosystem Function,
and Ecosystem Services. Front Sustain Food Syst. 4(September):1–13.
Thrift BD, Mosley JC, Brewer TK, Roeder BL, Olson BE, Kott RW. 2008. Prescribed sheep
grazing to suppress spotted knapweed on foothill rangeland. Rangel Ecol Manag. 61(1):18–
25.
Tiscornia G, Jaurena M, Baethgen W. 2019. Drivers, process, and consequences of native
grassland degradation: Insights from a literature review and a survey in Río de la Plata
grasslands. Agronomy. 9(5):8–12.
Vesk PA, Westoby M. 2001. Predicting plant species’ responses to grazing. J Appl Ecol.
38(5):897–909.
Wang Q, Li Y, Wang Y. 2011. Optimizing the weight loss-on-ignition methodology to
quantify organic and carbonate carbon of sediments from diverse sources. Environ Monit
Assess. 174(1–4):241–257.
Wang X, Vandenbygaart AJ, McConkey BC. 2014. Land management history of Canadian
grasslands and the impact on soil carbon storage. Rangel Ecol Manag. 67(4):333–343.
Whitehead D. 2020. Management of grazed landscapes to increase soil carbon stocks in
temperate, dryland grasslands. Front Sustain Food Syst. 4(October):1–7.
doi:10.3389/fsufs.2020.585913.
Wickham, H. 2016. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New
York.
Wikeem BM, Mclean A, Quinton D, Bawtree A. 1993. An overview of the forage resource
and beef production on Crown land in British Columbia. Can J Anim Sci. 73(4):779–794.
Wikeem B, Wikeem S. 2004. The grasslands of British Columbia. Kamloops, BC:
Grasslands Conservation Council of BC.
Witzgall K, Vidal A, Schubert DI, Höschen C, Schweizer SA, Buegger F, Pouteau V, Chenu
C, Mueller CW. 2021. Particulate organic matter as a functional soil component for persistent
soil organic carbon. Nat Commun. 12(1):1–10.
Wooley SC, Smith B, King C, Seastedt TR, Knochel DG (2011) The lesser of two weevils:
physiological responses of spotted knapweed (Centaurea stoebe) to above- and belowground
herbivory by Larinus minutus and Cyphocleonus achates. Biocontrol Sci Technol 21:153–
170

71

Zhai X, Zhang Y, Wang K, Chen Q, Li S, Huang D. 2018. Grazing effects on the nutritive
value of dominant species in steppe grasslands of northern China. BMC Ecol. 18(1):1–9.

72

3. CHAPTER 3
TESTING THE EFFICACY OF TARGETED CATTLE GRAZING TO SUPPRESS
SPOTTED KNAPWEED (CENTAUREA STOEBE) IN SEMI-ARID RANGELANDS

Introduction
Spotted knapweed (Centaurea stoebe) is biennial to short-lived perennial plant, first
introduced to British Columbia in 1893 through contaminated soil used as ship ballast
(Sheley et al. 1998). Further introductions were made through contaminated alfalfa seed
(Maddox 1982). First-year spotted knapweed plants form a leafy rosette and second-year
plants ‘bolt’ then become reproductive, producing an average of 1,000 seeds
that remain viable for up to eight years (Davis et. al 1993; Gayton and Miller 2012). In
heavily infested areas, spotted knapweed seed density has been reported to be as high as
5,000 to 40,000 seeds m2 (Schirman 1981; Sheley et al. 1998; Benzel et al. 2009).
Furthermore, plants can grow up to 1.5 meters in height and live up to nine years (Boggs and
Story 1987; Sheley et al. 1998), however, the average spotted knapweed lifespan is 3-5 years
(Watson and Renny 1974; Boggs and Story 1987). Spotted knapweed spreads primarily by
seed, yet they can also develop two to six
lateral stems that produce numerous
reproductive branches (Sheley et al. 1998;
Martin et al. 2014).
Spotted knapweed has several features
that give it a competitive advantage over
native grassland species. Unlike their
graminoid counterparts, spotted
knapweed lacks extensive underground
root structures (Figure 3.1). Instead,
knapweed has a large taproot that can
access deeper soil water and nutrients

Figure 3.1. Comparison of spotted knapweed's
taproot and the fibrous root structure of
bluebunch wheatgrass (Photo credits: Marci
Hess and Sage Grouse Initiative).
73

(Kelsey and Locken 1987), and reduce soil stability leading to erosion (Sheley et al. 1998;
Rinella et al. 2001; Hook et al. 2004; Knochel et al. 2010). Fraser and Carlyle (2011) also
discovered that large patches of spotted knapweed may uptake soil nitrogen and carbon at
higher rates, therefore decreasing the availability of these elements for native plant species.
The effects of spotted knapweed invasion are felt province-wide, where its presence
negatively impacts recreation values, wildlife habitat, and agricultural practices (Maxwell et
al. 1992; Rinella et al. 2001; Gayton and Miller 2012; Lloyd et al. 2019). As of 2021, spotted
knapweed is estimated to have infested 27,802 hectares of land in British Columbia (B.C.)
(IAPP 2021). This estimate is very conservative, however, as the provincial Invasive Alien
Plant Program (IAPP) may not capture infestation area data on private lands or locations with
limited road access.
As a consequence of knapweed’s rapid spread, B.C.’s agriculture producers are struggling
with reduced pasture values (Benzel etl al. 2009; Magnin et al. 2016; Mosley et al. 2016) and
the added expense and labour associated with chemical and mechanical control methods
(Griffith and Lacey 1991; DiTomaso 2000; Marchetto et al. 2021). There are also limitations
to which treatments can be applied based on the landscape type, climate, proximity to water,
infestation size, and available resources. Therefore, to successfully restore an area infested
with, and degraded by, spotted knapweed it’s recommended to integrate a combination of
treatment options, such as mechanical, cultural, chemical and biological controls (Sheley and
Jacobs 1997; DiTomaso 2000; van Wilgen et al. 2001; Lake and Minteer 2018). Utilizing
invasive plants as forage is becoming an increasingly popular and effective tool to help
control invasive plants and limit their spread (Coffey 2007; Bohnert et al. 2014; Mosley et al.
2016; Rinella and Bellows 2016; Marchetto et al. 2021).
Targeted grazing involves developing specific vegetation management objectives, such as
reduced invasive plant cover, then altering the duration, timing, and intensity of grazing to
achieve those said objectives (Bailey et al. 2019; Marchetto et al. 2021). For instance, it’s
ideal to ‘target’ an invasive plant when it’s most susceptible to defoliation, and at a time
when the impacts on non-target species is minimal (Olson et al. 1997; Bailey et al. 2019).

74

Domestic livestock were originally believed to avoid spotted knapweed and it was considered
poor quality forage (Harris and Cranston 1979; Maddox 1982; Kelsey and Mihalovich 1987;
Kennett et al. 1992). This idea was supported by the fact that knapweed contains a bittertasting compound called cnicin that was believed to reduce the plants’ palatability (Kelsey
and Locken 1987; Ganguli et al. 2010). However, spotted knapweed is now recognized as
nutritious forage (Thrift et al. 2008; Ganguli et al. 2010) and is comparable to that of native
pasture and agronomic grasses (Burritt and Hart 2014).
Recent studies have found that sheep, goats, and cattle readily eat knapweed in all life stages
(Henderson et al. 2012; Rinella and Bellows 2016; Ganguli et al. 2010). In Montana’s
bluebunch-fescue dominated rangeland, Thrift et al. (2008) achieved an average spotted
knapweed utilization rate, or aboveground biomass removal, of 46% when grazed by sheep
in July. Furthermore, research conducted by Henderson et al. (2012) found that grazing
knapweed in late July during spotted knapweed’s late bud to early flowering stage reduced
the production of viable seeds by nearly 100%. The sheep and cattle used in the Henderson et
al. (2012) study also appeared to prefer the knapweed over the grasses. Therefore, targeted
cattle grazing has the potential to be an effective tool to help suppress spotted knapweed.
In theory, targeted cattle grazing could deplete knapweed growth and the soil seed bank,
while also providing quality forage for cattle in mid to late summer. Grazing spotted
knapweed at a point when it’s susceptible to defoliation may also reduce its root mass (Olson
et al. 1997), therefore depleting the plants’ nutrient reserves over time.
When applying targeted grazing as an invasive plant management tool, it’s important to
consider the post-treatment effects, and next steps towards developing a range that’s more
resilient to invasion by non-native species. Gallandt (2006) recommends that rangeland
invasive plant management strategies should be long-term, plant-specific, and integrate
methods of depleting the weed seed bank.
Grazing can greatly influence the soil seed bank. For instance, seed banks can be depleted of
their diversity and size as a result of heavy grazing (Soloman et al. 2005). Seed banks clearly
play a critical role in restoration research, although the literature on soil seed bank

75

composition in Western Canadian rangelands is limited, especially in response to targeted
grazing strategies (Clements et al. 2007; Plue and Hermy 2012; Dobb and Burton 2013, Pyle
2018). Based on the existing literature, there is a need to develop management strategies to
control spotted knapweed that also promote the restoration of native plant communities after
knapweed treatment.
Research Objectives
There were two knowledge gaps I address in this study:
1) Will targeted cattle grazing help suppress the invasive plant spotted knapweed in B.C.’s
native, semi-arid rangelands?
2) Can the existing soil seed bank help restore invaded, native plant communities in semiarid rangeland following targeted grazing?
To address these knowledge gaps, I conducted a 9-week field experiment and a 16-week soil
seed bank study in TRU’s research greenhouse. The field experiment was conducted
in the same location as described in Chapter 2. Therefore, no site description is included in
Chapter 3.
First of all, I hypothesized that targeting spotted knapweed would reduce the number of seed
heads and seeds produced when compared to the control, ungrazed plants. Secondly, targeted
plants would exhibit lower above and belowground biomass in response to the grazing
treatment. Lastly, there would be fewer spotted knapweed seeds observed within the targeted
soil seed bank when compared to the control, ungrazed enclosure.

76

Materials and Methods
Field Experiment
In July 2019, two 50 x 50 m electric fence enclosures – one targeted and one ungrazed
control – were constructed at Drum Lake Pasture (Figure 3.2). Prior to July 2019, enclosures
were grazed under the 2017-2021 Chutter Ranch LTD. range use plan and followed a springfall grazing rotation (Table 3.1). Vegetation surveys conducted in late June 2020
demonstrated that the targeted and control enclosures exhibited similar spotted knapweed
cover values at 16.5% and 16.1%, respectively (Appendix 3.1)
Table 3.1. Description of the grazing schedule for Drum Lake pasture as outlined in the 2017
- 2021 RUP prepared by Chutter Ranch Ltd.
Even years

Odd years

Pasture name

Drum Lake

Drum Lake

Period of use

Sept 19 - 28

June 1 - 10

500 C/C

500 C/C

No. and class of livestock

20 Bulls

Figure 3.2. Map illustrating the targeted (grazed) and control (ungrazed) treatment
enclosures and knapweed sampling transects located at Drum Lake pasture in Merritt, BC.

77

On August 4th, 2020 the targeted grazing enclosure was grazed at the height of spotted
knapweed flowering by ten cow-calf pairs for a 24-hour period, equivalent to 0.8 animal unit
months (AUMs) ha-1. To help with distribution, cattle were supplied with four 40-gallon
stock troughs filled with water, where one trough was placed in each corner of the enclosure.
Ethics Statement
This research was carried out under Thompson Rivers University Animal Care and Use
Protocol No. 101899.
Spotted Knapweed Plant Traits & Response to Targeted Grazing
On June 22nd 2020, 20 single-stemmed and bolting spotted knapweed plants were selected in
each treatment enclosure along two randomly-placed 50 m transects (Figure 3.2).
Approximately every 2 meters, a knapweed plant closest to the transect was permanently
tagged using a landscape pin with flagging tape. Field measurements were recorded weekly
and included the number of buds and flowers and plant height. Plants were also monitored
for regrowth, or the production of lateral stems and rosettes. All tagged knapweed plants
were harvested on August 31st, prior to seed dispersal, by digging a hole deep enough to
capture the taproot and carefully removing the entire plant. Plants were stored in paper bags,
then brought to the Thompson Rivers University research greenhouse.
Harvested knapweed samples were oven-dried at 65°C for 48 hours and mass of the root and
aboveground material was recorded on an analytical balance. The number of seed heads per
plant were recorded, then seed heads were removed and stored at room temperature in paper
bags until seeds could be counted. Knapweed seeds were sorted using a 10x magnification
lens into the following categories: 1) mature (fully developed, black, hard seed coat), 2)
immature (light brown or gray, hard seed coat), 3) doughy (soft seed coat, white or
translucent) (Benzel et al. 2009), and 4) damaged seeds (Figure 3.3). Damaged seeds were
identified by having evidence of feeding from biological control insects. In this study,

78

doughy and damaged knapweed seeds were assumed to be unviable based on findings from
Benzel et al. 2009.

Figure 3.3. A) Dissecting spotted knapweed seed heads and sorting the knapweed seeds
by stages of maturity. B) A closer look at the mature spotted knapweed seeds collected.
Soil Seed Bank Sampling
Soil seed bank sampling was conducted from September 28-30th, 2020. Within the 50 m by
50 m control and targeted grazing enclosures, twelve 1 m2 quadrats were placed in a
sampling grid (Figure 3.4). Within each 1 m2 frame, twelve soil cores with a diameter of 5
cm and a depth of 5 cm (98.17 cm3) were removed using a bulb planter. The soil cores within
each 1m2 quadrat were pooled to form a composite sample containing 1178 cm3 or 1.178 L of
soil, a total of approximately 14.14 L of soil per grazing enclosure (24 composite samples or

79

288 soil cores total). A buffer zone of two meters was maintained between the fence and the
sampling grid to avoid edge effects.

Figure 3.4. Soil seed bank sampling design within each: A) 50 m by 50 m treatment
enclosure, B) 1m2 sampling quadrat frame.
Greenhouse Study
Soil seed bank samples were placed in a chest freezer for three months to experience a cold
stratification period, then grown in the greenhouse. Perforated germination trays (28 x 56 x
5.5 cm) were divided into two sections with landscaping fabric then filled with a 2 cm layer
of sterile sand. Each composite soil sample was sieved with 4 mm mesh to remove rocks and
plant materials, then spread out into a labeled germination tray; soil samples were allowed to
air dry for two days prior to sieving. Samples were randomly assigned to a germination tray,
then exposed to a 16-h day and 8-h night regime, a relative humidity of 30%, and a
temperature setting of 20-25°C, as recommended by Plue and Hermy (2012).
Emerging seedlings were identified, counted, then removed (Figure 3.5). Unknown seedlings
were transplanted into potting soil and left to grow until identification was possible. Rough
and Idaho fescue seedlings were indistinguishable and were therefore grouped as ‘fescue
species’. If no seeds germinated within a two-week period, the soil was gently turned over
using a trowel. After 12 weeks, a 1% gibberellic acid solution was applied in an attempt to
stimulate the growth of remaining, dormant seeds (Finkelstein et al. 2008). The greenhouse
pod was monitored and maintained for external weed growth to prevent seeds from
80

dispersing and contaminating the germination trays. The germination experiment was
terminated after 16 weeks. Soil seed bank density was then calculated and converted to the
number of seeds/m2.

Figure 3.5. The soil seed bank germination trays set up in the greenhouse, and a selection of
seedlings identified in the seed germination experiment: A) Microsteris gracilis, B)
Centaurea stoebe rosette, C) Collinsia parviflora and D) Lupinus sericeus.
Statistical Analysis
Knapweed root and shoot weights were compared using independent samples T-test, where
shoot weights had to be log transformed to meet normality and equal variance assumptions.
Root-shoot ratio values did not meet the requirements for a parametric test following
transformation attempts, therefore a non-parametric Mann-Whitney U test was used.
After transforming the raw data, the number of seed heads and the knapweed seeds at all
stages of maturity (mature, intermediate, doughy, and damaged) did not pass the Shapirowilk tests for normality. Therefore, the non-parametric equivalent of an independent samples
T-test, the Mann-Whitney U test was used to compare knapweed seed characteristics
between the targeted and control plants.
Mean and standard error values presented in subsequent tables are from untransformed data.

81

Results
Spotted knapweed root weight passed passed the Shapiro-Wilks normality test (p = .382) and
Levene’s test of homogeneity of variances (p = .068). Log transformed knapweed shoot
weight passed the Shapiro-Wilks normality test (p = .303) and Levene’s test of
homogeneity of variances (p = .112). For the remaining knapweed plant trait data, the MannWhitney U test statistics and significance values are summarized in Table 3.2.
Knapweed Response to Targeted Grazing
I observed that 75% of the tagged and targeted knapweed plants were grazed, where 10% of
grazed plants exhibited regrowth, or the formation of one rosette post-grazing. Altogether,
185 knapweed seed heads were dissected and 2,456 knapweed seeds at all stages of maturity
were counted (mature = 1,191; intermediate = 329; doughy = 734; damaged = 202).
Table 3.2. Summarized mean spotted knapweed plant trait values (+/- 1 SE), in addition to
the mean knapweed seeds (+/- 1 SE) grouped by seed maturity. Mann-Whitney U and
independent sample t-test statistics are displayed. Bold p-values denote significance between
the control and targeted knapweed plant variables (α = .05).

Variable

Mean +/- 1 SE
Targeted
Control

Statistic

p

Seed Heads

1.5 ± .47

7.8 ± .73

U = 383.5

<.001

Shoot weight (g)

.93 ± .11

2.1 ± .24

T = -4.652

<.001

Root weight (g)

.48 ± .05

.54 ± .05

T = -.811

0.422

Root-shoot ratio

.57 ± .08

.29 ± .02

U = 63.5

<.001

Mature

5.6 ± 2.97

54.0 ± 9.76

U = 376.0

<.001

Intermediate

1.5 ± .59

14.9 ± 3.12

U = 348.0

<.001

Doughy

5.1 ± 1.83

31.7 ± 3.73

U = 376.0

<.001

Damaged

1.4 ± .79

8.7 ± 1.41

U = 351.0

<.001

Knapweed Seeds

The targeted grazing treatment reduced the number of knapweed seed heads by over 80%
(Table 3.2). Furthermore, knapweed shoot weight was reduced by 56% in the targeted plants
82

respective to the control, ungrazed plants (Table 3.2). Before and after measurements
displayed that an average of 12.9 cm of plant material (by height) was removed from the
targeted plants as a result of the grazing treatment.

No differences were observed in the root mass between the targeted and control spotted
knapweed plants (Table 3.2). Moreover, the root-shoot ratio was calculated to be smaller in
the control plants compared to the targeted knapweed plants (Table 3.2).
Cattle readily consumed spotted knapweed in early August at the flowering stage and
reduced the number of mature (p < .001) and intermediate (p < .001) seeds by 89.7 and 90%
respective to control, ungrazed plants (Table 3.2).
Soil Seed Bank
Altogether, 740 seeds germinated and 17 species were detected in our sampling of the upper
five centimeters of the soil seed bank (Table 3.3). Of those species, seven were introduced
and ten were native to B.C.. The targeted and control enclosures had average seed densities
of 1186 seeds/m2 and 1304 seeds/m2, respectively (Table 3.3).
Targeted grazing did not appear to reduce the number of spotted knapweed seeds in the seed
bank, as samples contained an average of 153.1 seeds/m2 compared to 146.1 seeds m/2 in the
control enclosure (Table 3.3). Moreover, introduced plant species made up 32% of the
targeted seed bank and 27% of the control, whereas the remaining 68% and 73% were native
species.

83

Table 3.3. Soil seed bank composition within the targeted and control enclosures at Drum
Lake pasture. Species status, the total number of observed seedlings, and the average seed
density (m2) for each species is presented.
Common name

Scientific name

Status

Total # seeds

Average seed
density (㎡)

Targeted Control Targeted Control
Cheatgrass
Fescue Species

Bromus tectorum
Festuca idahoensis
Festuca campestris

Introduced
Native

29

4

100.9

13.9

56

86

194.8

299.2

Junegrass

Koeleria macrantha

Native

80

20

194.8

69.6

Kentucky bluegrass

Poa pratensis

Introduced

41

47

142.6

163.5

Sandberg's bluegrass

Poa secunda spp secunda Native

58

16

201.8

55.7

Cut-leaved daisy

Erigeron compositus*

Native

1

0

3.5

0

Holboell's rockcress

Arabis holboellii

Native

1

3

3.5

10.4

Silky lupine

Lupinus sericeus

Native

3

2

10.4

7.0

Slender phlox

Microsteris gracilis

Native

5

1

17.4

3.5

Small-flowered blueeyed Mary

Collinsia parviflora

Native

36

142

125.2

494

Spotted knapweed

Centauria stoebe

Introduced

44

42

153.1

146.1

Spring draba

Draba verna*

Native

2

0

7.0

0

Tall tumble mustard

Sisymbrium altissimum*

Introduced

4

0

13.9

0

Introduced

1

3

3.5

10.4

Western tansy mustard Descurainia sophia
Yarrow

Achillea millefolium

Native

3

0

10.4

0

Yellow alyssum

Alyssum alyssoides

Introduced

1

6

3.5

20.9

Yellow salsify

Tragopogon dubius

Introduced

0

3

0

10.4

Totals:

17 species

365

375

1186.3

1304.6

* Species that were not detected in vegetation surveys.

Junegrass, fescue species, Sandberg’s bluegrass, Kentucky bluegrass, and spotted knapweed
were the dominant species detected within the seed bank. Fescue was more dominant within
the control enclosure, whereas Sandberg’s and Junegrass were dominant within the targeted
enclosure. Cheatgrass was also more prevalent within the targeted enclosure at 100.9
seeds/m2 compared to just 13.9 seeds/m2 in the control.

84

Discussion
This study demonstrated that targeted cattle grazing has the potential to suppress spotted
knapweed growth and seed production within B.C.’s semi-arid rangelands. Our research
findings were consistent with Rinella et al. (2001), Benzel et al. (2009), and Mosley et al.
(2016), where one defoliation period during spotted knapweed’s bud to flowering period is
effective in reducing seed production. Mature spotted knapweed seeds are not formed until
post-flowering (Watson and Kenny 1974), which at Drum Lake was observed in mid to late
August. Despite this, a few individual knapweed plants may have been more advanced and
matured earlier than others. Therefore, Wallander et al. (1995) and Benzel et. al (2009)
recommend to quarantine livestock for 7 days after feeding on knapweed plants in full
flower.
There is little, to no research on how the digestive tract of cattle influences knapweed seed
viability. Wallander et al. (1995) fed rams, also a ruminant animal, approximately 5,000
mature spotted knapweed seeds and measured the recovery and viability of the seeds
following digestion. They recovered 17% of the digested seeds, and percent viability ranged
from 0-22% (Wallander et al. 1995). No viable seeds were detected two days after the
knapweed seed feeding (Wallander et al. 1995). Therefore, the digestive tract of ruminant
animals, such as sheep and cattle, have the potential to reduce the viability of mature spotted
knapweed seeds. By repeating our experiment, cattle could be quarantined, and their manure
collected, to observe whether knapweed seeds could be recovered. If seeds were detected
within cattle manure, it’s critical to test what percentage of the seeds would remain viable
and for how long. That way, a more refined quarantine duration could be determined.
Furthermore, minimal regrowth was observed in the grazed spotted knapweed plants, with
just 10% of the plants forming a rosette following grazing. Targeting spotted knapweed
during mid-summer in semi-arid range makes plant recovery difficult (Mosley et al. 2016), as
B.C.’s interior rangelands typically have very low soil moisture (Battigelli et al. 2003). In our
experiment, knapweed plants could have been monitored into the fall when soil moisture
increases, instead of being harvested in late August. This would have provided us with a
clearer picture of knapweed survival and regrowth following targeted grazing.
85

With any targeted grazing strategy, it is important to ensure there are no adverse effects to
the existing native plant community, wildlife, or to soil structure and function (Coffey 2007;
Bailey et al. 2019; Marchetto et al. 2021). Future research should investigate the long-term
effects of targeting spotted knapweed (> 5 years) on plant community dynamics, soil
structure and function, and the soil seed bank.
It is also worth testing different types of livestock, as they exhibit different feeding
preferences, forage intakes, and behaviours (Fraser 2004; Coffey 2017; Pauler et al. 2020).
For instance, a study conducted in western Montana by Henderson et al. (2012) found that
grazing with sheep and cattle sequentially suppressed spotted knapweed growth. Both
animals also consumed more knapweed than grasses during spotted knapweed’s late bud –
early flowering stage (Henderson et al. 2012). Therefore, the timing of spotted knapweed
defoliation, in addition to the type of grazing animal is important to consider.
It is well understood that cattle benefit from grazing taller forage plants; when forage plants
are shorter, they are less accessible and we may begin to see reduced forage intake and body
condition over time (Fraser 2004; Pauler et al. 2020). Spotted knapweed’s rosette stage is
leafy and palatable, but is also lower to the ground making this stage more difficult for cattle
to graze. Whereas the later knapweed growth stages form woody, more fibrous stems that the
cattle tend to avoid (Pauler et al. 2020). Whereas browsers, such as goats, can access shorter
plants and they are better able to strip leaves from stems, and can easily chew and digest
woody stems (Coffey 2007). Future research should apply multiple-species grazing in B.C.’s
semi-arid rangeland, where two or more species of grazers are used (Coffey 2007).
This study also lacked replication, and should ideally be repeated in lower, middle and upper
grassland sites. To do this, lower elevation grassland sites may need to be grazed earlier in
the growing season due to warmer temperatures, less moisture, and more rapid plant growth
and maturity (Wikeem and Wikeem 2004). A study investigating how spotted knapweed
growth variables (including bud formation, flower development and seed set) vary by season
and elevation would be very useful in our understanding of targeted grazing applications in
B.C. rangelands.

86

Additionally, this study should be repeated in areas with varied degrees of infestation, for
instance mild, moderate, and severe. Our study site had a relatively ‘light’ infestation (Thrift
et al. 2008) of spotted knapweed, where the average percent cover amongst all pastures at
Drum Lake was approximately 16%. It is important to determine whether targeted cattle
grazing would be successful in areas with higher knapweed cover, or if favourable plant
community shifts, such as improved cover of perennial bunchgrasses, would occur.
Soil Seed Bank
Targeted cattle grazing significantly reduced the number of seeds, at all stages of maturity,
produced by spotted knapweed. However, knapweed seeds within the seed bank did not
differ from the control, ungrazed enclosure. The results contradict Olson et al. (1997), where
they observed more viable knapweed seeds within ungrazed areas than in the sheep-grazed
areas after three consecutive summers of grazing. However, this study was limited in that the
targeted enclosure was grazed for just one season. Ideally, this grazing experiment would
continue for multiple years and be monitored frequently to determine if the knapweed seed
bank would, in fact, also become depleted over time. Unfortunately, invasive plant
suppression and removal is not a short-term project, and often requires many years of active
treatment and management to achieve results. Based on the existing, literature, there is
evidence to support that grazing will suppress the seed production of spotted knapweed over
time (Olson et al. 1997; Thrift et al. 2008; Benzel et al. 2009; Henderson et al. 2012).
One reason that the targeted and control enclosure displayed similar seed densities could be
that knapweed seeds have a long viability, up to seven years (Davis et. al 1993). There was
also a presence of seed-feeding biological control insects (Larinus minutus) and knapweed
seed head gall flies (Urophora spp.) on site, which may have contributed to a decline in
knapweed seed numbers (Seastedt et al. 2007). Granivores, or small mammals and birds
predating upon the knapweed seeds, could have also played a role in the knapweed seed
densities seen in the experiment (Bernards and Morris 2017). Many seed caches of spotted
knapweed were observed on site (Figure 3.6), and it brings to question the role of granivores
and biological control insects in suppressing, or spreading, spotted knapweed seeds.

87

Figure 3.6. Spotted knapweed seed caches, presumably formed from
the small mammal population on site.
Conclusions & The Future of Targeting Grazing
In conclusion, this research has demonstrated the effectiveness of targeted grazing in semiarid rangelands. Integrating targeted grazing into rangeland weed management plans may
help improve range condition and help create more productive, invasive free rangelands in
B.C.’s Southern Interior.
Instead of complete knapweed eradication being the end goal, we should learn to adapt,
manage, and utilize the invasive plants that are present, if at all possible. The climate is
changing and semi-arid rangelands are becoming more susceptible to plant invasion
(Clements and Ditommaso 2011; Clements and Jones 2021). As land managers, it is
important to consider – to what degree of invasive plant infestation is ‘tolerable’? In other
words, when do invasive plants, such as spotted knapweed, begin to have adverse effects on
the surrounding vegetation and soils?
It is also unclear which threats may be introduced into our rangelands in the future. For
instance, there are species listed on the provincial Early Detection Rapid Response (EDRR),
such as Medusahead (Taeniatherum caput-medusae) and North-Africa grass (Ventenata
dubia) that actively being monitored for, and pose risks to invade rangelands (PoBC 2021).

88

Therefore, it is critical that we continue to promote adaptive and integrated land management
now, and into the future.
In conclusion, perhaps it is time to rethink the standard management practice of moving
cattle into mid-high elevation forested understories to graze in summer months. Instead, it
may be possible to provide cattle with adequate forage and graze areas of range that have
moderate to severe infestations of spotted knapweed, or other palatable invasive plants. More
research should be conducted on the nutritive values of other prevalent rangeland invasive
plant species found in B.C., and targeting plants other than spotted knapweed should be
explored at a large scale.

89

REFERENCES
Bailey DW, Mosley JC, Estell RE, Cibils AF, Horney M, Hendrickson JR, Walker JW,
Launchbaugh KL, Burritt EA. 2019. Synthesis paper: Targeted livestock grazing:
prescription for healthy rangelands. Rangel Ecol Manag. 72(6):865–877.
Battigelli JP, McIntyre GS, Broersma K, Krzic M. 2003. Impact of cattle grazing on
prostigmatid mite densities in grassland soils of southern interior British Columbia. Can J
Soil Sci. 83(5):533–535.
Benzel KR, Mosley TK, Mosley JC. 2009. Defoliation timing effects on spotted knapweed
seed production and viability. Rangel Ecol Manag. 62(6):550–556.
Boggs KW, Story JM. 1987. The population age structure of spotted knapweed (Centaurea
maculosa) in Montana. Weed Sci. 35(2):194–198.
Bohnert DW, Sheley RL, Falck SJ, Nyman AA. 2014. Knapweed hay as a nutritional
supplement for beef cows fed low-quality forage. Rangel Ecol Manag. 67(2):219–223.
Burritt B, Hart RA. 2014. Nutritional value and toxins in various noxious weeds. Utah State
University: Logan, Utah. 15 p.
Burton CM, Burton PJ. 2003. A manual for growing and using seed from herbaceous plants
native to the Northern Interior of British Columbia. Smithers, British Columbia. 168p.
Clements DR, Krannitz PG, Gillespie SM. 2007. Seed bank responses to grazing history by
invasive and native plant species in a semi-desert shrub-steppe environment. Northwest Sci.
81(1):37–49.
Clements DR, Ditommaso A. 2011. Climate change and weed adaptation: Can evolution of
invasive plants lead to greater range expansion than forecasted? Weed Res. 51(3):227–240.
doi:10.1111/j.1365-3180.2011.00850.x.
Coffey L. 2007. Targeted grazing: A natural approach to vegetation management and
landscape enhancement. Launchbaugh KL, editor. Centennial, CO: American Sheep Industry
Association.
Davis ES, Fay PK, Chicoine TK, Lacey CA. 1993. Persistence of spotted knapweed
(Centaurea maculosa) Seed in Soil. Weed Sci. 41(1):57–61.
DiTomaso JM. 2000. Invasive weeds in rangelands: Species, impacts, and management.
Weed Sci. 48(2):255–265.
Dobb A, Burton SL. 2013. British Columbia rangeland seeding manual. Abbotsford: British
Columbia, BC Ministry of Agriculture. 282 p.

90

Fraser, D. A. 2004. Factors influencing livestock behaviour and performance. Forest
Practices Branch, British Columbia Ministry of Forests, Victoria, B.C. Rangeland Health
Brochure 8.
Gallandt ER. 2006. How can we target the weed seedbank? Weed Sci. 54(3):588–596.
Ganguli AC, Hale MB, Launchbaugh KL. 2010. Seasonal change in nutrient composition of
spotted knapweed and preference by sheep. Small Rumin Res. 89(1):47–50.
Gayton D, Miller V. 2012. Impact of biological control on two knapweed species in British
Columbia. Ecosyst Manag. 13(3):1–14.
Henderson SL, Mosley TK, Mosley JC, Kott RW. 2012. Spotted knapweed utilization by
sequential cattle and sheep grazing. Source Rangel Ecol Manag. 65(3):286–291.
Hobbs RJ, Hallett LM, Ehrlich PR, Mooney HA. 2011. Intervention ecology: Applying
ecological science in the twenty-first century. Bioscience. 61(6):442–450.
Hook PB, Olson BE, Wraith JM. 2004. Effects of the invasive forb Centaurea maculosa on
grassland carbon and nitrogen pools in Montana, USA. Ecosystems. 7(6):686–694.
[IAPP] Invasive Alien Plant Program. 2021. Invasive alien plant program application: A
database for invasive plant data in B.C. B.C. Ministry of Forests, Lands, Natural Resource
Operations and Rural Development. https://maps.gov.bc.ca/ess/hm/iapp/
Kelsey RG, Locken LJ. 1987. Phytotoxic properties of cnicin, a sesquiterpene lactone from
Centaurea maculosa (spotted knapweed). J Chem Ecol. 13(1):19–33.
Kelsey RG, Mihalovich RD. 2007. Nutrient Composition of spotted knapweed (Centaurea
maculosa). J Range Manag. 40(3):277.
Kennett GA, Lacey JR, Butt CA, Olsonrutz KM, Haferkamp MR. 1992. Effects of
defoliation, shading and competition on spotted knapweed and bluebunch wheatgrass. J
Range Manag. 45(4):363.
Kimball S, Schiffman PM. 2003. Differing effects of cattle grazing on native and alien
plants. Conserv Biol. 17(6):1681–1693.
Lacey, C. A., J. R. Lacey, P. K. Fay, J. M. Story, and D. L. Zamora. 1995. Controlling
knapweed in Montana rangeland. Montana State University Cooperative Extension Service
Circ. 311. 17 p.
Lake EC, Minteer CR. 2018. A review of the integration of classical biological control with
other techniques to manage invasive weeds in natural areas and rangelands. BioControl.
63(1):71–86.

91

Lloyd D, Iverson K, Mackenzie K, Richardson H, Jeffries L. 2019. Laurie Guichon
Memorial Grasslands Interpretive Site: A baseline inventory. Kamloops, BC. Grassland
Conservation Council of British Columbia. 98 p.
Maddox DM. 1982. Biological control of diffuse knapweed (Centaurea diffusa) and spotted
knapweed (C. maculosa). Weed Sci. 30(1):76–82. doi:10.1017/s0043174500026217.
Marchetto KM, Wolf TM, Larkin DJ. 2021. The effectiveness of using targeted grazing for
vegetation management: a meta‐analysis. Restor Ecol. doi:10.1111/rec.13422.
Martin LM, MacDonald NW, Brown TE. 2014. Native plant establishment success
influenced by spotted knapweed (Centaurea stoebe) control method. Ecol Restor. 32(3):282–
294.
Maxwell JF, Drinkwater R, Clark D, Hall JW. 1992. Effect of grazing, spraying, and seeding
on knapweed in British Columbia. J Range Manag. 45(2):180.
Mosley JC, Frost RA, Roeder BL, Mosley TK, Marks G. 2016. Combined herbivory by
targeted sheep grazing and biological control insects to suppress spotted knapweed
(Centaurea stoebe) . Invasive Plant Sci Manag. 9(1):22–32.
Müller-Schärer H, Schaffner U, Steinger T. 2004. Evolution in invasive plants: Implications
for biological control. Trends Ecol Evol. 19(8):417–422. doi:10.1016/j.tree.2004.05.010.
Olson BE, Wallander RT, Lacey JR, Olson BE, Wallander RT, Lacey JR. 1997. Effects of
Sheep Grazing on a Spotted Knapweed-Infested Idaho Fescue Community. 50(4):386–390.
Pauler CM, Isselstein J, Suter M, Berard J, Braunbeck T, Schneider MK. 2020. Choosy
grazers: Influence of plant traits on forage selection by three cattle breeds. Funct Ecol.
34(5):980–992. doi:10.1111/1365-2435.13542.
Plue J, Hermy M. 2012. Consistent seed bank spatial structure across semi-natural habitats
determines plot sampling. J Veg Sci. 23(3):505–516.
Province of British Columbia (PoBC). [Internet]. 2021. Early detection and rapid response
(EDRR). Available from: https://www2.gov.bc.ca/gov/content/environment/plants-animalsecosystems/invasive-species/priority-species/priorityplants#Early%20detection%20and%20rapid%20response%20(EDRR)
Pyle LA. 2018. Influence of management and disturbance history on germinable seed bank
composition and legume recruitment in Alberta’s Central Parkland and Dry Mixedgrass
prairie. University of Alberta.
Rinella MJ, Jacobs JS, Sheley RL, Borkowski JJ. 2001. Spotted knapweed response to season
and frequency of mowing. J Range Manag. 54(1):52–56. doi:10.2307/4003527.
Rinella MJ, Bellows SE. 2016. Evidence targeted grazing benefits to invaded rangelands can
increase over extended time frames. Rangel Ecol Manag. 69(3):169–172.
92

Rutter SM. 2010. Review: Grazing preferences in sheep and cattle: Implications for
production, the environment and animal welfare. Can J Anim Sci. 90(3):285–293.
Seastedt TR, Knochel DG, Garmoe M, Shosky SA. 2007. Interactions and effects of multiple
biological control insects on diffuse and spotted knapweed in the Front Range of Colorado.
Biol Control. 42(3):345–354.
Sheley RL, Jacobs JS, Carpinelli MF. 1998. Distribution, biology, and management of
diffuse knapweed (Centaurea diffusa) and spotted knapweed (Centaurea maculosa). Weed
Technol. 12(2):353–362.
Sheley RL, Svejcar T, Maxwell B. 1996. A theoretical framework for developing
successional weed management strategies on rangeland. Weed Technol. 10(4):766–773.
Solomon TB, Snyman HA, Smit GN. 2006. Soil seed bank characteristics in relation to land
use systems and distance from water in a semi-arid rangeland of southern Ethiopia. South
African J Bot. 72(2):263–271.
Thrift BD, Mosley JC, Brewer TK, Roeder BL, Olson BE, Kott RW. 2008. Prescribed sheep
grazing to suppress spotted knapweed on foothill rangeland. Rangel Ecol Manag. 61(1):18–
25.
Toledo D, Sanderson M, Spaeth K, Hendrickson J, Printz J. 2014. Extent of Kentucky
bluegrass and its effect on native plant species diversity and ecosystem services in the
Northern Great Plains of the United States. Invasive Plant Sci Manag. 7(4):543–552.
Watson, A. K., and A. J. Renny. 1974. The biology of Canadian weeds. Centaurea diffusa
and C. maculosa. Canadian Journal of Plant Science 54: 687–701
Wilgen B van, Richardson D, Higgins S. 2001. Integrated control of invasive alien plants in
terrestrial ecosystems. L Use Water Resour Res. 5:6.
Wikeem B, Wikeem S. 2004. The grasslands of British Columbia. Kamloops, BC:
Grasslands Conservation Council of BC. http://doi.wiley.com/10.2307/1931227.

93

4. CHAPTER 4
RESEARCH SUMMARY AND MANAGEMENT IMPLICATIONS

Research Summary and Recommendations
This research assessed whether intensive grazing practices can create productive, invasivefree rangelands in the Southern Interior of British Columbia. Results from Chapter 2
demonstrated that it’s best to exercise caution if implementing management-intensive grazing
(MiG) in native, semi-arid rangelands. There continues to be discrepancy as to whether MiG
is beneficial to rangeland health over extensive or more conventional grazing methods
(Briske et al. 2011; Teague et al. 2013). The results of the MiG treatment demonstrated
promising increases in total and native grass cover, reduced bare ground, and improved
diversity indices when compared to baseline conditions. However, similar results were seen
within the extensive grazing enclosure. Therefore, future studies should attempt to
disentangle the influence of grazing deferment or period of rest between grazing applications
on vegetation and soil characteristics. A longer study duration may have painted a clearer
picture of grazing effects.
There were also multiple challenges identified that may prevent the use of MiG on large land
areas. First, implementing MiG on expansive crown or rural private land would be difficult
due to the lack of water and fencing developments in natural range. Not only are these
structures expensive, but they require a lot of maintenance. Solar panels and a battery, or an
alternative an energy source is required for the electric portable fencing. Water needs to be
readily accessible and available for livestock, and either needs to be transported in via truck,
water pump from an adjacent water source, or built water lines. The time and labour required
to manage distant pastures and adjust fencing is very consuming. Livestock also need to be
trained to get comfortable with the electric fencing, and being confined in smaller enclosures.
Lastly, terrain can be difficult to navigate in some areas, such as steep slopes and rocky
outcrops.

94

That being said, MiG may be more applicable and successful when used on small-scale
ranching operations, with smaller livestock herds. However, more research is needed to
explore the benefits of MiG, especially if implementing multiple grazing periods throughout
the year (i.e. shorter deferment periods), and grazing earlier in the spring when preferred and
vulnerable bunchgrasses are actively growing.
Chapter 3 describes the first applied field study in B.C. which tested targeted cattle grazing to
control spotted knapweed. Our study demonstrated that targeting spotted knapweed with
cattle at the late bud – flowering stage in late July can help reduce the number of seeds
produced and dispersed. However, we also observed that knapweed cover and biomass
increased over a 2-year duration, possibly due to a compensatory growth response. It’s
recommended to supplement the targeted grazing treatment described in this study with
alternative forms of weed management. This could look like:
1) Spot spraying surviving knapweed plants with a broadleaf, residual herbicide
immediately after the targeted grazing treatment. This may help improve control of
the ungrazed knapweed plants. This could be followed with an application of native
seed in the fall, prior to snowfall.
2) An early fall application of a residual, broadleaf herbicide with an active ingredient
such as 2,4-D or picloram (commercial names Grazon, Tordon 22K). This would help
target plants that did not get grazed, kill rosettes that emerge in the fall, and suppress
rosette and plant growth in the spring. Apply a native seed mixture in the early spring
immediately after snowmelt.
3) Monitoring for existing populations of biological control insects, then release insects
to initiate a population or supplement existing biocontrol populations.
Within semi-arid rangelands, it is important to practice adaptive management. Grassland
ecosystems in particular can change drastically from year to year due to factors such as
precipitation, periods of drought, fire, or previous year’s management practices (Iverson
2004; GCC 2017; Pauler et al. 2020).
When I think of the overall scope of this research, I consider whether we should aim to
‘restore’ ecological communities to their original condition, or to an alternate (still
95

functioning) state (Hobbs et al. 2011). When developing ecological restoration strategies, it’s
important to consider and define goals prior to initiating the project. What features of the
degraded system would you like to see improved? For instance, in an overgrazed natural
grassland, is increased native plant cover and productivity a goal? Increased soil organic
matter and nutrient cycling? Within a grassland ecosystem invaded with an undesirable
species, what will the restored plant community look like? Invasive plant infestations tend to
persist over time. What is a tolerable percent cover for the area, and what will help maintain
this over time? These are all questions to consider when practicing and applying integrated
rangeland management.
Further Research Opportunities
Practicing integrated land management, or applying multiple forms of weed control at once,
may aid in the fight against invasive plants and promote rangeland restoration (Coffey 2007;
Mosley et al. 2016; Lake and Minteer 2018). One method is biological control insects, or
biocontrol. In British Columbia, there are numerous biocontrol insects that have been
introduced to help control spotted knapweed invasions (Gayton and Miller 2012). Field
observations suggest that there are established populations of the root feeding knapweed
weevil (Cyphocleonus achates), the knapweed seed feeding weevil (Larinus minutus), and
the knapweed gall fly (Urophoa sp.) (Gayton and Miller 2012; Seastedt and Knochel 2021).
Biocontrol do not completely eradicate the target plant species; instead, following their
introduction, biocontrol causes a decline in the target plant density and eventually reach an
equilibrium; therefore, a ‘tolerable’ level of host plant density is achieved, which no longer
causes the severe economic, social, or ecological impacts it once did (Louda et al. 2003;
Barratt et al. 2017).
Moreover, the effectiveness of seeding trials following targeting spotted knapweed is still
uncertain, specifically in semi-arid or dryland environments. Future research should
investigate different combinations and application rates of native and agronomic seed
following targeted grazing to determine if they are successful. Providing information on the
existing soil seed bank is important so that it can supplemented with specialized seed
applications and treatments.

96

Reseeding is common tool used in restoring native plant communities, however, sourcing
preferred seeds can be expensive to purchase and apply at landscape scales (Burton and
Burton 2003). Deciding to spread non-native or agronomic seed may also diminish the social
and ecological values of native grasslands (Dobb and Burton 2013).
Therefore, for seed treatments to be successful, there are multiple considerations and
requirements to be made. The timing and rate of seed application must be decided upon, and
the seed mix must be suitable for the site’s climate. Moreover, the seed must come into
contact with mineral soil, have access to light, receive adequate moisture, and avoid
predation from birds and small mammals (Dobb and Burton 2013). Attempting to seed semiarid range is particularly difficult due to low and variable rainfall; therefore, seed
establishment in dryland sites may be as low as 1% (Shakelford et al. 2021). B.C. ranching
operators also may not have the capacity or resources available to make reseeding successful;
moreover, seeding may not be economically feasible due to the added expenses of the seed,
equipment, and labour.
Despite the challenges that come with seed applications, research shows that reseeding with
native species immediately following spotted knapweed control may encourage native plant
establishment and provide competition against the knapweed (Martin et al. 2014). Seed
mixes that are selected specifically to compete against spotted knapweed should be
investigated and developed. It would also be worthwhile to test the combined effects of
targeted grazing and biocontrol on spotted knapweed density, seed bank and productivity,
compared to biocontrol or grazing alone.

97

REFERENCES
Barratt BIP, Moran VC, Bigler F, van Lenteren JC. 2018. The status of biological control and
recommendations for improving uptake for the future. BioControl. 63(1):155–167.
Briske DD, Sayre NF, Huntsinger L, Fernandez-Gimenez M, Budd B, Derner JD. 2011.
Origin, persistence, and resolution of the rotational grazing debate: Integrating human
dimensions into rangeland research. Rangel Ecol Manag. 64(4):325–334.
Coffey L. 2007. Targeted grazing: A natural approach to vegetation management and
landscape enhancement. Launchbaugh KL, editor. Centennial, CO: American Sheep Industry
Association.
Dobb A, Burton SL. 2013. British Columbia rangeland seeding manual. Abbotsford: British
Columbia, Ministry of Agriculture.
Gayton D, Miller V. 2012. Impact of biological control on two knapweed species in British
Columbia. Ecosyst Manag. 13(3):1–14.
Grasslands Conservation Council of British Columbia (GCC). 2017. Managing BC
grasslands: an overview. Kamloops, BC. 22p.
Iverson K. 2004. Grasslands of the southern interior. Victoria, BC: Ministry of Sustainable
Resource Management. 6p.
Lake EC, Minteer CR. 2018. A review of the integration of classical biological control with
other techniques to manage invasive weeds in natural areas and rangelands. BioControl.
63(1):71–86.
Louda SM, Arnett AE, Rand TA, Russell FL. 2003. Invasiveness of some biological control
insects and adequacy of their ecological risk assessment and regulation. Conserv Biol.
17(1):73–82.
Mosley JC, Frost RA, Roeder BL, Mosley TK, Marks G. 2016. Combined herbivory by
targeted sheep grazing and biological control insects to suppress spotted knapweed
(Centaurea stoebe). Invasive Plant Sci Manag. 9(1):22–32.
Pauler CM, Isselstein J, Suter M, Berard J, Braunbeck T, Schneider MK. 2020. Choosy
grazers: Influence of plant traits on forage selection by three cattle breeds. Funct Ecol.
34(5):980–992.
Seastedt TR, Knochel DG. 2021. A 20-year evaluation of successes with biological control of
spotted knapweed (Centaurea stoebe) in Colorado. Biol Control. 159(April):104631.

98

Shackelford N, Paterno GB, Winkler DE, Erickson TE, Leger EA, Svejcar LN, Breed MF,
Faist AM, Harrison PA, Curran MF, et al. 2021. Drivers of seedling establishment success in
dryland restoration efforts. Nat Ecol Evol. 5(9):1283–1290.
Teague R, Provenza F, Kreuter U, Steffens T, Barnes M. 2013. Multi-paddock grazing on
rangelands: Why the perceptual dichotomy between research results and rancher experience?
J Environ Manage. 128:699–717.

99

Appendix 1 – Vegetation survey plant list from Drum Lake Pasture in Merritt, B.C.

100

Appendix 2 – The ‘Par + Den 5’ method described in the modified fractionation
protocol received from the Summerland Research and Development Centre (SuRDC),
one of Agriculture and Agri-food Canada’s research centres.
Sieving method for separation of fractions with different turnover rates
(based on Par+Den5 method described in Poeplau et al. 2018)

Fractionation protocol
1. Spread samples into metal pans. Pull larger root bits or living plants out of the
sample. Leave in greenhouse to air-dry. If samples are really wet and/or full of lots of
roots, let the sample partially air dry and then sieve (2 mm) when still fairly moist to
remove roots (this should prevent contamination of the light fraction with dried out
bits of freshly killed roots).
2. Otherwise, wait until samples are dry, and pass through a 2 mm sieve.
3. Store air-dried samples in plastic bags at room temperature.
4. Number and record weights of all clean beakers and pans that will contain soil or
POM fractions NOTE: use a 4 decimal scale for all measurements.

101

5. Place ~50 g of soil (record wt. ) in a 250 mL Erlenmyer flask, add 150mL distilled
water and seven (5 mm) glass beads, swirl to mix and shake overnight (16 hours) on
orbital shaker setting 4.5 (225 rpm).
6. Organize pre-weighed beakers as follows:
a. F>50 (fraction 50-200 um, rinsed from 50 um sieve)
b. Pan (F<50 fraction rinsed from bottom catch pan) (600mL beaker)
c. POM>50 (floated POM panned from F>50)

7. Stack sieve catch pan, and 50 um sieve. Swirl flask and remove any debris on flask
walls with stir rod and pour contents onto 50 um sieve. Rinse the flask onto the sieve.
Remove glass beads with tweezers. Tap and stir soil on sieve with plastic utensils
(spatulas or basting brushes) to help push liquid through.
8. Lightly rinse material on the 50 um sieve and sieve walls with spray bottle to encourage
finer soil particles into the pan below. Rinse backside of sieve into the pan being careful
to angle the sieve so as not to lose fractionated material on top into the pan below.
9. With scraping and minimal water, rinse the fractions from the pan (Pan F<50) and the
sieve (F>50 + POM) into the appropriate labelled beakers placed inside an aluminum
catch pan. Minimize spillage. Rinse soil from utensils and spillage from catch pans into
appropriate fractions and then rinse all equipment once finished. At this stage, do NOT
separate the POM>50 fraction from the F>50 mineral soil fraction.
*NOTE: to ensure the majority of organic matter is removed from the 50um sieve, use
high pressure distilled water tap to force the OM into a corner to be easily removed
with spatula/more rinsing after thoroughly rinsing all soil/residue into beaker.
10. Label beakers with their respective sample # and fraction. Stick labels to the rim of
beakers to avoid any sticker residue on the beaker. Stick labels to the rims of beakers
so they can be used for the plastic vials after drying.
11. Dry all fractions at 65C until the POM in the F>50 fraction is dry before starting the
following step:

102

Re-floatation with NaI
(only if significant organic matter is observed in the dried F>50 sample)
i)
scrape/loosen all contents of the beaker
ii)
based on the amount of POM present, re-float the contents of the beaker with
50mL (lower POM amount) up to 100mL of NaI solution for higher POM
content.
iii)
shake 5 minutes (vigorous stirring for 1 min. can be substituted for shaking)
iv)
after a short settling time (~2min) the floating material is POM and the sunken
material is the mineral fraction (F>50). Using a spoon, remove as much POM
that floats or that can be visually identified. Add this to the POM>50 tin. (place
directly onto filter paper?)
v)
using a pipette, collect as much of the remaining POM as possible onto the filter
paper. There is no need to minimize the amount of liquid collected in the tin as
everything will be vacuum filtered.
vi)
using a No. 4 Whatman paper on a Buchner funnel under vacuum, place the
contents of the POM tin on the filter, then proceed to carefully pour through the
filter much of the remaining NaI and floated POM from the beaker (without
losing any sunken mineral fraction). Rotate the beaker as you pour to help
separate the POM from the mineral soil and collect POM from beaker edges.
Collect the pure NaI for re-use.
vii)
rinse the POM with DH2O by flooding using a larger rinse bottle and then
smaller bottle to help contain the material. Rinse until liquid is clear and any
residual yellow colouring on filter paper is very faint. Clean off sides of funnel
and encourage POM towards center of the filter paper using a fine spray bottle
(DH2O) and spatula. Rinse POM off utensils onto paper. Shut off vacuum. Fold
and remove the filter paper (now containing the rinsed POM) from the Buchner
funnel and rinse (with DH2O) the contents off the filter paper into a clean, preweighed plastic beaker (POM>50). Rinse any residual from the funnel into
beaker.
viii) Repeat these rinsing steps with the mineral fraction remaining in the beaker to
remove any NaI.
ix)
placed rinsed mineral fraction into a clean beaker and re-sieve through the 50um
sieve and pan stack to further separate into proper fractions using the steps
above. Add the contents of sieve and pan to the appropriate fractions.
x)
re-dry all fractions at 65C for 48-72 hours. Cool and take final weights.
12. Grinding: use grinding mill for all samples. Alternatively, POM can be ground with
mortar/pestle.
13. If sample sizes after grinding are very small, place samples directly into tin or silver
foil capsules used for analysis, crimp top and replace in labelled vial.
14. Set aside unfractionated soils for further analysis:
i.
DOC, by Xiying
ii.
Specific surface area, by Kirsten/Andy
iii.
Exchange cations/Fe/Al, by Kirsten
103

Equipment:
250 mL Erlenmeyer flasks (40)
Dispenser (capable of 150 mL)
Scale
Weighing scoops
5mm glass beads (300)
Tweezers
Shaker
Pan and 50um sieves
Plastic beakers, 200mL and 600mL
Small tin pans (for POM)
Scoopula and
Disposable transfer pipettes
NaI solution, d = 1.7 g/mL (Soil Sampling and Methods of Analysis (Martin R. Carter) p. 400)
Method development
Feb 12/20
In an attempt to streamline the process above we considered floating the POM right after the
initial sieving (instead of first drying the F>200 with the POM in it, then floating). This
meant adding NaI to a slurry of wet soil and POM. We also attempted to vacuum filter the
F>200+POM fraction first to remove as much water as possible. In both cases, attempting to
float the POM in NaI without drying the fraction first was difficult and much mixing of the
different fractions was observed. Conclusion is that oven drying the F>200+POM is
important for the efficacy of the floatation process.

104

Feb 20/20
Upon observing the similar 13C signatures of the coarse (F>200) and fine (F>50) sand
fractions it is likely not useful to have these as separate fractions. For this reason, the 200um
sieve could be eliminated. The advantages are:
i)
ii)
iii)

All sand combined into one fraction
Reduced sample numbers
Flotation with NaI will now help in capture POM>50, not just POM>200

105

Appendix 3 – Summarized vegetation survey data for the targeted and control
treatment enclosures at Drum Lake Pasture.
Mean % Cover ( +/- 1 SE)
Common name
Scientific name
Dandelion
Taraxacum officinale
Meadow death camas
Toxicoscordion venenosum
Flixweed (Tansy mustard)
Descruainia sophia
Holboell's rockcress
Arabis holboellii
Lemonweed
Lithospermum ruderale
Nuttall's pussytoes
Antennaria parviflora
Parsnip-flowered buckwheat
Eriogonum heracleoides
Purple fleabane species
Erigeron spp.
Salsify
Tragopogon dubius
Small-flowered blue-eyed Mary Collinsia parviflora
Silky Lupine
Lupinus sericeus
Slender phlox
Microsteris gracilis
Spotted knapweed
Centauria stoebe
Timber milkvetch
Astragalus miser
Tiny trumpet
Collomia linearis
Yarrow
Achillea millefolium
Yellow alyssum
Alyssum alyssoides
Bluebunch wheatgrass
Pseudoroegneria spicata
Kentucky bluegrass
Poa pratensis
Junegrass
Koeleria macrantha
Sandbergs bluegrass
Poa secunda spp secunda
Cheatgrass
Bromus tectorum
Idaho fescue
Festuca idahoensis
Rough fescue
Festuca campestris
Ground Parameters
Bare Ground
Rock
Litter
Crust
Dung

Functiona
l group
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Forbs
Grasses
Grasses
Grasses
Grasses
Grasses
Grasses
Grasses

Status
Introduced
Native
Introduced
Native
Native
Native
Native
Native
Introduced
Native
Native
Native
Introduced
Native
Native
Native
Introduced
Native
Introduced
Native
Native
Introduced
Native
Native

Targeted
.9 ± .53
0
1.6 ± .73
.6 ± .31
.2 ± .20
0
7.3 ± 4.2
0
.5 ± .34
0
8.4 ± 2.67
.3 ± .15
16.5 ± 2.95
4.0 ± 1.0
.1 ± .10
16.1 ± 2.58
.7 ± .42
8.0 ± 1.61
2.2 ± 1.13
7.7 ± 2.14
4.3 ± .84
.1 ± .10
1.7 ± 1.16
0

Control
1.1 ± .41
.7 ± .52
.2 ± .20
.5 ± .34
0
4.6 ± 3.0
9.0 ± 3.7
.7 ± .52
1.6 ± .48
.1 ± .10
19.5 ± 6.52
.1 ± .10
16.1 ± 5.85
.3 ± .30
.5 ± .31
18.2 ± 3.62
.6 ± .43
0
3.0 ± 1.34
3.9 ± .77
4.2 ± 1.13
0
11.9 ± 2.35
1.8 ± 1.5

6.9 ± 1.66
4.5 ± .85
13.5 ± 1.2
3.3 ± 1.34
1.2 ± .1.0

3.9 ± 1.35
2.6 ± 1.43
21 ± 4.65
3.5 ± 1.0
.1 ± .10

106