EVALUATING THE GRAZING RESPONSE INDEX OF THREE KEY FORAGE
SPECIES IN THE SOUTHERN INTERIOR OF BRITISH COLUMBIA BY USING
MULTIPLE CLIPPING TREATMENTS TO DETERMINE IMPACT ON PLANT
VIGOUR

By
VANESSA VOLPATTI

Bachelor of Natural Resources Sciences, Thompson Rivers University, 2011

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF
MASTER OF SCIENCE IN ENVIRONMENTAL SCIENCES

in the Faculty of Science
in
Environmental Science
Thesis Supervisor:
Dr. Wendy Gardner
Committee Members:
Dr. Lauchlan Fraser
Dr. Reg Newman
Lavona Liggins

2013
Thompson Rivers University
© Vanessa Volpatti, 2013

ACKNOWLEDGEMENTS

I would like to thank my supervisor, Dr. Wendy Gardner for all of her guidance
and support that she gave me through this process and always kept me going. My
committee members also provided me with a lot of help and instruction: Lavona Liggins
for all her coordination with Agriculture and Agri-Food Canada and particularly during
my first field season of work helping me find my study sites and especially finding extra
helping hands to assist me. Reg Newman for all his help with my statistical design and
analysis as well as helping to locate study sites. Lauchlan Fraser for helping decide the
best ways to use my large amounts of data and providing valuable insight and direction.
I thank my research assistants Katie Clark and Kayla Holtslag for their many
hours of monotonous work counting tillers and measuring tiller heights and making my
research and work fun and interesting, rain or shine; Erika Tessier, Tori Waites and
Mahesh Khadka for filling in whenever I needed extra assistance.
My family and friends have been a great support network throughout the duration
of my Master’s. My parents Reno and Sandra Volpatti for giving me a lot of emotional,
physical and financial support; my siblings, Gina, Trevor and Craig for their emotional
and physical support. Chris Maciel for his continued encouragement and support that
never let me forget I could accomplish all of my goals and that anything worth doing is
worth doing right.
A final thank you to Agriculture and Agri-Food Canada for supporting and
funding this project.

ii

DEDICATION
This paper is dedicated to my two late Grandfathers, Joseph Volpatti and Fred Chantler,
both of whom were kind hearted, hardworking men who each gave me a piece of
themselves. For my Grandpa Volpatti who spent his whole life dedicated to farming,
agriculture and his family and gave me those same passions that he once had. For my
Grandpa Chantler, who was a soft spoken man with a strong connection with nature and
the great out-doors and gave me a keen interest and respect for the natural world.

iii

ABSTRACT
In British Columbia, many practices related to rangeland management are not
working as effectively as they used to due to fluctuating environmental factors. The
Grazing Response Index (GRI) is a tool which was developed in Colorado, USA to help
rangeland managers and producers evaluate the effects of grazing in a current year by
integrating management and climate factors which relate directly to growing conditions.
To determine if this tool could be used in the Southern Interior of British Columbia, in
conjunction with range condition assessments and range health assessments, the
effectiveness of the GRI was determined by comparing the responses of three key forage
species to various levels of clipping: clipped once at 40% or 70% or clipped three times
at 40% or 70% removal of biomass. Results varied by species: bluebunch wheatgrass was
impacted greater by intensity of clipping rather than frequency, rough fescue showed
interactions between frequency and intensity while pinegrass results were variable.
Though results varied by species, the GRI scoring for each species response was
considered appropriate though it was conservative in its scoring of the more severe
treatments. I conclude that the GRI could be a beneficial tool for annual range
management in the Southern Interior to supplement long term management tools.

iv

TABLE OF CONTENTS
ACKNOWLEDGEMENTS ............................................................................................................. ii
DEDICATION ................................................................................................................................ iii
ABSTRACT ................................................................................................................................... iv
TABLE OF CONTENTS ................................................................................................................. v
LIST OF TABLES ......................................................................................................................... vii
LIST OF FIGURES ....................................................................................................................... vii
LIST OF APPENDIX FIGURES..................................................................................................... x
CHAPTER 1: GENERAL INTRODUCTION ................................................................................ 1
1.1 BRITISH COLUMBIA RANGELANDS .............................................................................. 1
1.2 RANGE MANAGEMENT AND PLANT GROWTH .......................................................... 3
1.3 RANGELAND MONITORING ............................................................................................ 7
1.4 THE GRAZING RESPONSE INDEX .................................................................................. 9
1.5 STUDY SITE ....................................................................................................................... 10
1.6 OUTLINE OF THESIS ........................................................................................................ 12
CHAPTER 2: EFFECT OF FREQUENCY AND INTENSITY OF CLIPPING ON
BLUEBUNCH WHEATGRASS (Pseudoroegneria spicata), ROUGH FESCUE (Festuca
campestris) AND PINEGRASS (Calamagrostis rubescens) IN THE SOUTHERN INTERIOR
OF BRITISH COLUMBIA............................................................................................................ 13
2.1 INTRODUCTION ............................................................................................................... 13
2.2 MATERIAL AND METHODS ........................................................................................... 18
Study sites .............................................................................................................................. 18
Experimental design............................................................................................................... 21
Measurements ........................................................................................................................ 22
2.3 RESULTS ............................................................................................................................ 26
Weather .................................................................................................................................. 26
Regrowth and soil moisture ................................................................................................... 30
Final biomass ......................................................................................................................... 34
Cumulative 2012 biomass ...................................................................................................... 36
Maximum tillers ..................................................................................................................... 38

v

Minimum tillers ..................................................................................................................... 40
2.4 DISCUSSION ...................................................................................................................... 45
Regrowth and soil moisture ................................................................................................... 45
Measurements of plant vigour................................................................................................ 48
2.5 CONCLUSION .................................................................................................................... 52
3.1 INTRODUCTION ............................................................................................................... 53
3.2 THE GRAZING RESPONSE INDEX ................................................................................ 53
Frequency............................................................................................................................... 54
Intensity ................................................................................................................................. 54
Opportunity ............................................................................................................................ 55
Creating a score...................................................................................................................... 56
3.3 EVALUATING THE GRAZING RESPONSE INDEX BASED ON PLANT VIGOUR
RESPONSE TO CLIPPING TREATMENTS ........................................................................... 57
Scoring of three key forage species by treatment .................................................................. 58
Evaluation conclusion ............................................................................................................ 62
3.4 FINAL CONCLUSION ....................................................................................................... 64
APPENDICES ........................................................................................................................... 73

vi

LIST OF TABLES

Table 1: Source of variation for one-way ANOVA. ......................................................... 25

Table 2: Source of variation for two-way ANOVA.......................................................... 25

Table 3: The GRI scoring values for frequency, intensity and opportunity. Adapted from
Reed et al. 1999................................................................................................................. 55

Table 4: The GRI scores for each treatment based on 2012 results.................................. 62

LIST OF FIGURES
Figure 1: Approximate area and location of British Columbia’s range resources (Wikeem
and Wikeem 2004). ............................................................................................................. 2

Figure 2: Aerial image of study sites. Bluebunch wheatgrass and rough fescue sites fall
within the Lac du Bois Grasslands Protected area boundaries, while the pinegrass site lies
within an unused pasture on Agriculture and Agri-Food Canada property (Google Earth
2013). ................................................................................................................................ 19

Figure 3: Climate normal and 2011 and 2012 weather data for bluebunch wheatgrass site.
a). Average high temperature (°C). b). Average low temperature (°C). c). Precipitation
(mm). Error bars on climate line indicate standard deviation of untransformed data. ..... 27

Figure 4: Climate normal and 2011 and 2012 weather data for rough fescue site. a).
Average high temperature (°C). b). Average low temperature (°C). c). Precipitation
(mm). Error bars on climate line indicate standard deviation of untransformed data. ..... 28

vii

Figure 5: Climate normal and 2011 and 2012 weather data for pinegrass site. a). Average
high temperature (°C). b). Average low temperature (°C). c). Precipitation (mm). Error
bars on climate line indicate standard deviation of untransformed data. * April climate
average only includes 8 years of data. .............................................................................. 29

Figure 6: Regrowth (cm) and soil moisture (VWC) of bluebunch wheatgrass, 5, 7, 9, 11,
and 13 days after removing 40% or 70% of total biomass. a.) Bluebunch wheatgrass
regrowth and soil moisture in 2011. b.) Bluebunch wheatgrass regrowth and soil moisture
in 2012. Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data................ 31

Figure 7: Cumulative regrowth (cm) and soil moisture (VWC) of rough fescue, 5, 7, 9,
11, and 13 days after removing 40% or 70% of total biomass. a.) Regrowth and soil
moisture of rough fescue in 2011. b.) Regrowth and soil moisture of rough fescue in
2012. Data in graph represents untransformed data and error bars are for untransformed
data while statistical analysis was done with transformed data. ....................................... 32

Figure 8: Cumulative regrowth (cm) and soil moisture (VWC) of pinegrass, 5, 7, 9, 11,
and 13 days after removing 40% or 70% of total biomass. a.) Pinegrass regrowth and soil
moisture for 2011. b.) Pinegrass regrowth and soil moisture for 2012. Data in graph
represents untransformed data and error bars are for untransformed data while statistical
analysis was done with transformed data.......................................................................... 34

Figure 9: Interactions between frequency and intensity for rough fescue final biomass (g).
Data in graph represents untransformed data and error bars are for untransformed data
while statistical analysis was done with transformed data................................................ 35

Figure 10: Final biomass (g) and cumulative biomass 2012 (g) clippings from 2012 of all
three plant species. a.) Bluebunch wheatgrass final biomass. b.) Rough fescue final
biomass. c.) Pinegrass final biomass. d.) Bluebunch wheatgrass cumulative biomass. e.)
Rough fescue cumulative biomass. f.) Pinegrass cumulative biomass. Letters denote a
significance level of P<0.05. Data in graph represents untransformed data and error bars
are for untransformed data while statistical analysis was done with transformed data. ... 37

viii

Figure 11: Interactions between frequency and intensity for rough fescue cumulative
biomass 2012 (g). Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data................ 38

Figure 12: Maximum tillers counts of bluebunch wheatgrass plants in 2012. Letters
denote significance levels (P<0.05). Data in graph represents untransformed data and
error bars are for untransformed data while statistical analysis was done with transformed
data. ................................................................................................................................... 39

Figure 13: Maximum tillers counts of pinegrass plants in 2012. Letters denote
significance levels (P<0.05). Data in graph represents untransformed data and error bars
are for untransformed data while statistical analysis was done with transformed data. ... 40

Figure 14: Minimum tiller counts for bluebunch wheatgrass. a.) Minimum tiller counts
for 2011. b.) Minimum tiller counts for 2012. Letters denote significance levels (P<0.05).
Data in graph represents untransformed data and error bars are for untransformed data
while statistical analysis was done with transformed data................................................ 41

Figure 15: Minimum tiller counts for pinegrass in 2012. Letters denote significance levels
(P<0.05). Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data................ 41

Figure 16: Bluebunch wheatgrass canopy height (cm) for 2011 and 2012 seasons. a.)
2011 canopy height. b.) 2012 canopy height. Letters denote significance levels (P<0.05).
Data in graph represents untransformed data and error bars are for untransformed data
while statistical analysis was done with transformed data................................................ 42

Figure 17: Rough fescue canopy height (cm) for 2011 and 2012 seasons. a.) 2011 canopy
height. b.) 2012 canopy height. Letters denote significance levels (P<0.05). Data in graph
represents untransformed data and error bars are for untransformed data while statistical
analysis was done with transformed data.......................................................................... 43

ix

Figure 18: Interactions between frequency and intensity of clipping in rough fescue
canopy height (cm) in 2011 and 2012. a.) Rough fescue canopy height 2011 interactions.
b.) Rough fescue canopy height 2012 interactions. Data in graph represents
untransformed data and error bars are for untransformed data while statistical analysis
was done with transformed data. ...................................................................................... 44

Figure 19: Pinegrass canopy height (cm) for 2011 and 2012 seasons. a.) 2011 canopy
height. b.) 2012 canopy height. Letters denote significance levels (P<0.05). Data in graph
represents untransformed data and error bars are for untransformed data while statistical
analysis was done with transformed data.......................................................................... 45

LIST OF APPENDIX FIGURES

Appendix 1: Regressions of stubble height (cm) versus plant biomass (g) for each
species. a.) Bluebunch wheatgrass. b.) Rough fescue. c.) Pinegrass. Untransformed data
used for regression. ........................................................................................................... 73

Appendix 2: Maximum tiller counts for bluebunch wheatgrass in 2011. Letters denote
significance (P<0.05). Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data................ 74

Appendix 3: Maximum tiller counts for rough fescue. a.) Rough fescue maximum tiller
count in 2011. b.) Rough fescue maximum tiller counts in 2012. Letters denote
significance (P<0.05). Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data................ 75

Appendix 4: Maximum tiller counts for pinegrass in 2011. Letters denote significance
(P<0.05). Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data................ 76

x

Appendix 5: Minimum tiller counts for rough fescue. a.) Rough fescue minimum tiller
count in 2011. b.) Rough fescue minimum tiller counts in 2012. Letters denote
significance (P<0.05). Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data................ 77

Appendix 6: Minimum tiller counts for pinegrass in 2011. Letters denote significance
(P<0.05). Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data................ 78

Appendix 7: Summary of significance values and means for all vigour measurements in
all three species. a.) Bluebunch wheatgrass. b.) Rough fescue. c.) Pinegrass. ................. 79

xi

CHAPTER 1: GENERAL INTRODUCTION

1.1 BRITISH COLUMBIA RANGELANDS
There are many land types within British Columbia (B.C.) that are considered
rangeland; such as grasslands, wetlands, forest and alpine ranges. Rangelands are
important forage sources for wildlife and livestock and provide a multitude of economic
and social benefits. Rangelands in British Columbia are most often found between the
Rocky Mountains in the east and the Coastal mountains in the west but also include the
Peace River zone in the northeast where the Great Plains extend from Alberta into British
Columbia (Fig. 1) (Bawtree 2005, Wikeem and Wikeem 2004). The spread of rangelands
across the province, with variations in climate, soils and topography, is represented by a
wide variety of distinct plant communities, which make rangelands in British Columbia
some of the most diverse rangelands in North America (Wikeem et al.1993). There are 14
distinct biogeoclimatic ecosystem classification (BEC) zones characterized within the
province, 11 of which make up the core of British Columbia’s range resources (Campbell
and Bawtree 1998). These zones are reflective of the broadest vegetation complexes and
regional climate and are named after one or two dominant species within the zone. For
example, the bunchgrass zone is dominated by bluebunch wheatgrass and rough fescue
which are bunchgrass species. Zones can further be broken down into subzones and
variants; it is at these levels where most management decisions are made. The diversity
and classification of British Columbia’s rangelands makes it apparent that management
of these zones, not only for rangeland resources but other values as well, can be difficult.
The Southern Interior is one of the driest and warmest regions in the province and
to discuss the rangelands within this region it is best to refer to them as forested zones
and open grasslands (Bawtree 2005). Approximately 7.6 million hectares of land in the
Southern Interior is considered rangeland (Mclean 1967, McLean 1972) most of which is
forested.

1

Figure 1: Approximate area and location of British Columbia’s
range resources (Wikeem and Wikeem 2004).

These forested zones produce less forage per hectare than open grasslands, but the
volume of forested rangeland is much greater; 80% of rangeland in B.C. is forested
(Wikeem et al.1993). Douglas-fir, ponderosa pine and engelmann spruce trees are the
climax dominant species that form the canopies in most of the forested rangelands in the
Southern Interior (Wikeem et al.1993). Some forested rangeland has been seeded to
improve forage on cut blocks (50,000 hectares in B.C. since 1960) often with agronomic
species; lodgepole pine is typically the tree species dominating the cut block canopies
(Wikeem et al. 1993). Based on the elevational gradient (which also results in a moisture
gradient), the grassland areas of the Southern Interior can separated into three distinct
zones of lower, middle and upper grasslands. The upper grasslands are in fact classified
as being in the grassland phase of the Interior Douglas-fir BEC zone as soil moisture is
great enough in this area to support tree growth (Ministry of Environment, Lands and
Parks 2000). Often, encroachment of trees into the upper grasslands can become a
management concern. Rangeland, regardless of type, is either public or private land.

2

Approximately 80% of forested rangeland and 10% of grassland rangeland is
owned by the provincial government (Campbell and Bawtree 1998) with livestock
producers often leasing the land under grazing licenses (10 years) or grazing permits (5
years) (Ministry of Forests, Lands and Natural Resource Operations, 2013a). Currently
there are grazing leases (21 years) in existence in some areas of the province, though no
new ones have been granted recently (Ministry of Forests, Lands and Natural Resource
Operations, 2013b). Provincial land makes up 60% of total forage requirements for
ranchers while the remaining 40% is gained from private rangeland and irrigated pasture
land (Wikeem et al. 1993). With approximately 410,600 head of cattle (bulls, cows,
heifers, steers and calves) in B.C. as of January 2013, and an industry that on average
produces an economic contribution of $600 million annually, continued due diligence and
improvement for management of range resources is crucial for economical sustainability
of B.C.s cattle industry (Statistics Canada 2013, B.C. Cattlemen’s Association 2013).

1.2 RANGE MANAGEMENT AND PLANT GROWTH
Range management is defined by the Society for Range Management as: “A
distinct discipline founded on ecological principles and dealing with the use of
rangelands and range resources for a variety of purposes. These purposes include use as
watersheds, wildlife habitat, grazing by livestock, recreation, and aesthetics, as well as
other associated uses.” (Society for Range Management 1989). The type of management
on rangelands is dependent upon the management objective. Rangeland management is a
process that is continually being altered and improved through research and application
of grazing management plans and rangeland assessments; these tools aid in reaching
desired production levels or goals while maintaining suitable levels of sustainable use.
Research relating to range management first began in the United States in the beginning
of the 1900s (Pyke et al. 2002) and, in the following years, it was decided research was
needed in matters affecting the range industry (such as plant community and integrity) as
ecological and environmental awareness caused public concern for proper management
of rangelands (Milroy and McLean 1980). This research over the past century has led to

3

the development of many different management strategies for the grazing (and well as
other uses) of rangelands.
Essentially, managing rangelands for grazing purposes is accomplished by
understanding the basic principles of defoliation and how plant vigour is impacted by
defoliation. Removal of photosynthetic material through grazing or other methods of
defoliation can result in reducing a plant’s capacity to form carbohydrates and proteins,
though this can be dependent upon the amount of material removed. (Holechek et al.
2011). The vigour of a plant is identified as characteristics of a plant which relate to the
ability of productivity, reproduction, and competition (Caldwell 1984). A plants ability to
quickly produce vegetative and reproductive shoots, store nutrient reserves and have
effective rooting systems are key indicators of vigour (Caldwell 1984). There are many
studies that support the fact that intensity and timing of grazing can significantly affect
the vigour and productivity of plants. This can further be impacted by the frequency of
defoliation as well as seasonal influences such as temperature and precipitation; with
precipitation being a very important factor in determining type and productivity of
vegetation in an area (Holechek et al. 2011). Depending upon the frequency, intensity, or
timing of defoliation, plants may utilize carbohydrate reserves to produce new
photosynthetic material in order to return to pre-defoliation photosynthetic rates
(Campbell and Bawtree 1998, Richards 1993). Root growth can also be affected by
defoliation as root elongation has been shown to slow or stop after 40-50% onetime
removal of biomass while fine roots may begin to die and decompose soon after
defoliation (Holechek et al. 2011, Richards 1993, Ryle and Powell 1975).
Frequency refers to the number of times a plant is defoliated in a growing season
and is related to duration, how long animals remain in a given area. Often, the greater
number of times a plant is defoliated, the more detrimental defoliation becomes (Ferraro
and Oesterheld 2002) as generally there is an inverse relationship between plant yield and
frequency of defoliation (Ellison 1960). A single grazing of a plant will often not lead to
severe, long term impacts on plant vigour since loss of old leaves can have less of a
negative impact than loss of newer leaves (Richards 1993) but consecutive grazing

4

occurrences will remove the new regrowth on previously grazed plants which can result
in loss of vigour (Briske 1986). Each subsequent defoliation event results in loss of new
regrowth causing greater energy expenditure of the plant to recover the lost
photosynthetic material (Briske 1986). Frequency can be impacted by intensity of
defoliation since multiple defoliations can be tolerated by most plants at low intensities;
however moderate to severe intensities with increased frequency will result in decreased
plant vigour (Richards 1993).
Intensity, the amount of photosynthetic material removed, is related to stocking
rate which is defined by Holechek et al. (2011) as the relationship between the number of
animals per grazing area over a period of time. What is considered to be light, moderate
and heavy grazing intensities varies upon the range type as each range type can differ in
precipitation and plant community composition (Holechek et al. 2011). On many
rangelands, light/conservative grazing is considered to be 40% or less removal of
biomass, moderate grazing usually is represented by 41-50% of biomass removed, while
heavy grazing is between 51-60% and severe grazing is greater than 60% removal of
biomass (Holechek et al. 2006). Light grazing can have positive impacts when compared
to ungrazed plants as grazing may result in more growth during the growing season and a
higher total biomass production by the end of the growing season (Holechek et al. 2006,
Ganskopp and Bedell 1981, Johnson 1956). Moderate grazing can be defined as “take
half leave half” (Holechek et al. 2011). Removing no more than 50% of the biomass
would not have any positive impacts on the grazed plant and aside from the recovery of
lost photosynthetic material, plant vigour would likely not be negatively impacted.
(Holechek et al. 2011, Reed et al. 1999). Heavy and severe grazing has negative impacts
on plant vigour that could require years of recovery (Briske et al. 2008, Blaisdell and
Pechanec 1949). The impact of intensity on a plant can vary by species and timing of
defoliation (growth stage). The removal of meristematic tissue can be more detrimental to
plant vigour than the removal of equal proportions of vegetative tissue (Richards 1993);
and removal of a significant portion of biomass often results in reduced photosynthetic

5

capacity (Briske et al. 2008), carbohydrate storage and in some cases, root growth
(Richards 1993, Campbell and Bawtree 1998, Ferraro and Oesterheld 2002).
Recovery of a plant after defoliation can depend on a variety of factors but
opportunity for growth (via resource availability and timing of defoliation) before
defoliation or regrowth after defoliation plays an important role. Most plants require a
sufficient amount of time to store energy during the growing season to maintain vigour;
this can be related to time of year that grazing occurs, and how long animals remain in a
pasture (time and timing) (Reed et al. 1999). The less time a plant has to recover from
defoliation, the more negative the effects of defoliation will likely be (Ferraro and
Oesterhel 2002). Grazing within certain times of the growing season can change how a
plant responds; most often grazing a plant during dormancy at a higher intensity will not
do as much damage compared to a plant that is actively growing or has its growing points
elevated (Holechek et al. 2011, Ferraro and Oesterheld 2002, Branson 1956). Grazing
plants in a vegetative stage often results in less damage as reproduction and carbohydrate
storage is not active at this point and growing points are typically not elevated enough for
grazing cattle to remove (depending upon the species) (Holechek et al. 2011, Branson
1956). There will be less opportunity for a plant to regrow when it is defoliated in a
reproductive stage (elevated growing points and low carbohydrate levels) when compared
to defoliation at a vegetative stage. It is the combination of intensity, frequency and
opportunity that influences how a plant responds to defoliation. It is important to
understand that grazing can modify plant vigour, growth and plant interactions with the
surrounding environment in a multitude of complex ways, all of which can influence the
over-all impact of grazing on plant vigour and growth (Hilbert et al.1981).
Plant species on rangelands will vary in their response to defoliation due to
different morphological and physiological characteristics (Holechek et al. 2011). In order
to manage range resources for long term sustainable livestock grazing, managers not only
need to understand the concepts of plant growth and the principles that determine
recovery after defoliation but also need to have a strong knowledge of the range areas
they are managing. Range use history, species composition, climate, soils and

6

disturbances are other factors that can influence plant productivity and vigour on
rangelands and these factors can change over time. Regardless of what may cause these
changes over time, it is important to know the effectiveness of management of range
resources and determine how the range has changed over a period of time.

1.3 RANGELAND MONITORING
In 1949, Dyksterhuis developed a quantitative procedure for rangeland monitoring
through evaluating range condition by measuring foliar cover of various plant species
labeled as increasers, decreases and invaders and comparing them to the climax
community (Dyksterhuis 1949, Briske et al. 2005). This procedure, known as the range
condition method, is based on successional theory developed by Clements in 1916
(Briske et al.2005). Depending upon the proportion of these species found, the range is
considered to be in “excellent, good, fair or poor” condition. To use this method it is
suggested that permanent transects be put in place and re-monitored at 2 year intervals
(Dykersterhuis 1949). This method considers what the rangeland should look like without
human interference as well as having the desired plant community (what managers would
like to see) (Dyksterhuis 1949). The range manager can adjust the management strategies
accordingly to ensure objectives are being met. The desired plant community is one that
has been decided upon to meet the needs and values of people with a vested interest in
the management of the land and is set in place by a management plan (Borman and Pyke
1994). This means the desired plant community may not always reflect what would
typically be seen on an undisturbed range site (climax plant community) and instead
economic considerations (forage production) may require the desired plant community to
be a different condition class (Dyksterhuis 1949).
The range condition model could only show change in a plant community along a
single axis, from poor to excellent condition, and couldn’t account for other changes in a
community such as invasive species, tree encroachment, fire, or other disturbances.
Another model was introduced which accounts for the dynamics of plant communities
and their abilities to have vegetation changes occur in more than one direction (Briske et

7

al. 2005). The state and transition model can represent vegetation change along several
different axes, some reversible and some not; these changes in vegetation can occur
quickly (fire) or very slowly (encroachment of trees into grasslands) and can be caused
by natural or anthropogenic events (Westoby et al. 1989). This theory acknowledges that
ecological processes such as weather variation, soil erosion and fire can cause shifts in
vegetation communities (Briske et al. 2005) and gives a better idea of overall rangeland
health, not just range condition. This has lead to a range monitoring method that not only
takes into account plant communities but other broader ecosystem properties (Briske et
al. 2005) and is used between specific time periods to assess change.
From these two main models, range condition class and state and transition,
different long term monitoring tools have been developed. In British Columbia, most past
range assessments were based on the Dyksterhuis (1949) range condition concept
(Campbell and Bawtree 1998). This method is still in use today but some modifications
have been made to the methodology with the main one being that seral stages (early, mid,
late and potential natural community) are used instead of condition classes. The state and
transition model is the basis for two range health assessment methods currently used in
BC, the British Colombia Ministry of Forest and Range Uplands Function Checklist
assessment and the Grassland Monitoring Manual for British Columbia (Newman et al.
2011). The Uplands Function Checklist measures the following parameters: hydrology
and soils, biotic/vegetation, erosion/deposition, and mineral cycle (Fraser 2009). The
Grasslands Monitoring Manual measures the following parameters: plant community
composition/structure, nutrient and hydrological cycle, site stability, and invasive plants
(Delesalle 2009). These assessments can be used to monitor plant community change
over time to observe a trend in range health. To observe a trend, accurate assessments of
rangeland health should be conducted at the start and the end of a definitive time period
(five years or more on native rangelands) (Vallentine 1990).
Regardless of what monitoring method is applied, the common methods of
assessment involve long term monitoring over a period of several years to observe trends.
Rangeland managers are often faced with limited time, money and personnel to return to

8

assess these permanent monitoring sites at regular intervals (Campbell and Bawtree
1998) and this can often result in the rangeland being negatively impacted without
detection. Short term (yearly) monitoring tools could prove to be a useful supplement to
the existing long term monitoring tools. Determining impacts of grazing and disturbance
on plant communities at the end of every growing season would provide opportunity for
adaptive management of rangelands allowing range managers and livestock producers to
plan ahead for future grazing.

1.4 THE GRAZING RESPONSE INDEX
As reviewed in a previous section, forage production in the BEC zones in the
Southern Interior is heavily influenced by annual weather patterns. While current
assessments tools are no doubt essential, combination with an annual grazing assessment
tool that can factor in weather patterns and climate would enhance the ability of range
managers to keep track of negative and positive impacts on range condition between
years of more comprehensive rangeland health assessments.
In recent decades, climate change has become more prevalent causing fluctuations
in weather patterns that influence plant growth which in turn has caused variances in how
plants respond under grazing pressures as growing conditions change. Water availability
in particular determines how well a plant grows before and after a grazing event (Rustad
2008). In British Columbia, it is expected that changes in precipitation may lead to
increases in drought occurrences and an increase of the average temperature by 0.5°C per
decade (Hamann and Wang 2006). The impacts of climate change on rangeland resources
and the recognition that changes in range condition are not always easily detected
between growing seasons. This indicates that something more is needed to aid land
managers in meeting their objectives between the longer periods of time when rangeland
condition or range health assessments are conducted. This is the reasoning for the
creation of the Grazing Response Index (GRI). This index is a tool that gives a numerical
score to rangeland after the end of the growing season of each grazing year. The scores
are based on the three principles of plant growth: frequency, intensity and opportunity

9

(Reed et al. 1999). Each principle is scored and then all three are totaled to give a final
GRI score. Based on the score given (positive, neutral or negative) the range manager can
make important decisions on next year’s grazing such as reducing stocking rates,
increasing cattle dispersal or changing the season of use (Reed et al. 1999). Essentially
the GRI is a supplemental management tool used on an annual basis to allow for adaptive
management of rangelands. It is a tool that may work within the plant communities of the
Southern Interior of British Columbia and would be beneficial to livestock producers and
land managers in the years between larger and more comprehensive rangeland condition
assessments and rangeland health assessments.

1.5 STUDY SITE
The wide variety of plant communities and range types within British Columbia
can make determining an area for study sites difficult. The Lac du Bois Grasslands
Protected area in Kamloops, British Columbia has provided many research opportunities
due to its distinct environmental gradient (moisture gradient due to precipitation and
elevation) which produces distinct plant communities (Ministry of Environment, Lands
and Parks 2000). The surrounding area also provides several distinct vegetation zones
common throughout the Southern Interior making it an ideal area to conduct research.
Livestock grazing, homesteading and recreation have been common place in this area for
the last 150 years and for a better part of the 1900s much of it was overgrazed by cattle
and horses (Campbell and Bawtree 1998, Ministry of Environment, Land and Parks
2000). Since 1976 rotational grazing has been in place in the pastures of Lac du Bois and
this, along with restricted motorized recreation, has allowed the area to begin recovering
from past use events (Ministry of Environment, Lands and Parks 2000).
The soils of Lac du Bois are most commonly Chernozems. The lower grasslands
tend to have dry, thin, well drained soils which make these areas sensitive to disturbance
and making the cryptogrammic crust particularly important in these areas for preventing
erosion (Ministry of Environment, Lands and Parks 2000). These areas also tend to have
lower species diversity and soils that are well drained compared to other areas (Ministry

10

of Environment, Lands and Parks 2000). As the elevation rises, soils contain more
gravely till which provides better moisture retention for plants and with increase plant
community biomass, darker chernozemic soils occur (Ministry of Environment, Lands
and Parks 2000).
Species such as bluebunch wheatgrass (Pseudoroegneria spicata) and big
sagebrush (Artemitia tridentata) are the dominant plant species found at the lower
elevations. With increasing elevations, big sagebrush becomes less dominant but still is
present in the middle grasslands, while bluebunch wheatgrass remains the dominant grass
species and needle and thread grass (Hesperostipa comata) becomes more prevalent.
Arrow leaf balsam root (Balsamorhiza sagittata), nodding onion (Allium cernuum),
mariposa lily (Calochortus macrocarpus) and Thompson’s paintbrush (Castilleja
thompsonii) are all common throughout the lower and middle grassland phases (Ministry
of Environment, Lands and Parks 2000).
Rough fescue (Festuca campestris) becomes the dominant grass species in the
upper grasslands while bluebunch wheatgrass can still be found on some sites and aspen
trees (Populus tremuloides) can be found in wetter areas (Ministry of Environment,
Lands and Parks 2000). A host of other flowering plants and some shrubs are found in
the upper grasslands as well, including arrow leaved balsam root, upland larkspur
(Delphinium nuttallianum), chocolate lily (Fritillaria lanceolata), sticky geranium
(Geranium viscosissimum) and few flowered shooting star (Dodecatheon pulchellum).
Forage production in these grassland zones can be as low as 110 kg/ha on the dry lower
grassland sites but up to 1300 kg/ha in the upper grassland areas (Wikeem et al. 1993,
Campbell and Bawtree 1998). This large difference in production is primarily due to soil
moisture which increases with elevation (Wikeem et al. 1993, Campbell and Bawtree
1998).
Just outside the upper borders of Lac du Bois is the forested zone where much
grazing still occurs. This area ranges from very open canopy forests to closed canopy
forests. Pinegrass (Calamagrostis rubescens) is the dominant grass species in this area.
Many other forbs and shrubs are abundant such as wild strawberry (Fragaria virginiana),

11

northern bedstraw (Galium boreale), tiger lily (Lilium columbianum), common red
paintbrush (Castellija miniata), soopolallie (Shepherdia canadensis), tall Oregon-grape
(Mahonia aquifolium) and common snowberry (Symphoricarpos albus). Forage
production varies depending upon species composition, precipitation averages and the
amount of canopy closure but it can range from being unproductive in mature stands, up
to 900 kg/ha in open canopy or cut block sites (Wikeem et al. 1993, Campbell and
Bawtree 1998).

1.6 OUTLINE OF THESIS
Chapter 2 will present the research conducted through manipulation experiments
in Lac du Bois Grassland Provincial Park and outlying area near Kamloops, British
Columbia. The study is based on three dominant plant species found in distinct plant
communities: bluebunch wheatgrass in the lower grassland, rough fescue in the upper
grassland and pinegrass in the Interior Douglas-fir zone. The main research of this study
involves clipping each species at specific frequency and intensity levels and determining
the impact on plant vigour. Chapter 3 takes the information from Chapter 2 and uses it to
determine if the Grazing Response Index is a short term monitoring tool that can aid in
the management of rangelands in the Southern Interior of British Columbia.

12

CHAPTER 2: EFFECT OF FREQUENCY AND INTENSITY OF
CLIPPING ON BLUEBUNCH WHEATGRASS (Pseudoroegneria spicata),
ROUGH FESCUE (Festuca campestris) AND PINEGRASS (Calamagrostis
rubescens) IN THE SOUTHERN INTERIOR OF BRITISH COLUMBIA
2.1 INTRODUCTION
Over the last century, studies involving the clipping of various grass species have
been a common way to simulate grazing pressure in a variety of plant communities.
Many studies have focused on the response of a plant to various levels of defoliation
(intensity), the number of times defoliation occurs (frequency), and the time (season) at
which the defoliation occurs. The information gained from these studies resulted in the
development of many useful tools for evaluating the effects of grazing on plant
communities and specific plant species. As more and more is learned, we continue to
create new management practices and refine others. This study focuses on the recovery of
plants after clipping at varying frequencies and specific intensity levels with
consideration for soil moisture and weather patterns. The three key grass species in the
Southern Interior: bluebunch wheatgrass (Pseudoroegneria spicata), rough fescue
(Festuca campestris), and pinegrass (Calamagrostis rubescens) are the focus of this
research as they make up the majority of grazed species within their respective plant
communities. Bluebunch wheatgrass has been the subject of many clipping studies while
fewer studies have focused on rough fescue and pinegrass.
Bluebunch wheatgrass has been the subject of many studies in the Pacific
Northwest region of the United States, while fewer studies have focused on the northern
communities of bluebunch wheatgrass in Canada, specifically in British Columbia (BC).
This species is an important forage source for livestock and wildlife and is found
throughout the grasslands of BC, but dominates in the lower and middle grasslands
(Ministry of Environment, Lands and Parks 2000, Wikeem and Wikeem 2004). In a study
by McLean and Wikeem (1985a) in the Southern Interior, bluebunch wheatgrass plants
were subjected to 10 different clipping treatments which varied in either the date and
frequency of clipping, or intensity of clipping. They found that clipping bluebunch

13

wheatgrass to a stubble height (amount of herbaceous material remaining after
defoliation) of 5 cm on a weekly basis between the end of April and May resulted in high
mortality, or significantly reduced vigour in the surviving plants at a lower grassland site.
At a similar lower grassland site, the same result was found when plants were clipped to a
stubble height of 5 cm on a weekly basis between mid May and late June. The clippings
described above occurred when bluebunch wheatgrass was in its boot stage (transition
from vegetative to reproductive) or early flowering stage, which is more susceptible to
injury (Quinton et al. 1982). At this time the plant has used much of its carbohydrate
storage, and would likely not have opportunity for regrowth before summer dormancy
when clipped to such levels (McLean and Wikeem 1985a). Weekly clipping at a 20 cm
stubble height, season long when compared to clipping only in the fall showed no
significant difference in plant health when compared to the control treatment whereas
weekly clipping to a stubble height of 15 cm and 10 cm from mid April to summer
dormancy resulted in moderate to heavy mortality at both sites (McLean and Wikeem
1985a). Clark et al. (1998) found that live basal area of bluebunch plants decreased when
the whole plant was clipped to a stubble height of 7.6 cm, compared to control plants
(unclipped) and plants that only had half of their biomass cut to 7.6 cm which both saw
an increase in live basal area.
The general trend of published research shows that early defoliation of bluebunch
results in limited negative effects on plant vigour as long as the plant has enough time to
regrow prior to summer dormancy but with later, more frequent, or higher intensity
defoliation, more deleterious effects occur (Blaisdell and Pechanec 1949, Mclean and
Wikeem 1985a, Quinton et al. 1982, Clark et al. 1998, Meays et al. 2000). Quinton et al.
(1982) showed that fall grazing when the plant is dormant will not cause injury to the
plant itself and will remove unpalatable stems, allowing for better quality forage the
following spring, however; Ndwala-Senyimba et al. (1971) and Wikeem et al. (1989)
found that removing the canopy structure of bluebunch wheatgrass can result in less rain
and snow capture and less protective litter which may result in reduced growth the
following year.

14

Rough fescue is another very important forage species in regions of BC and is the
dominant grass species in the upper grasslands, but is also found the middle grasslands
and occasionally in the lower grasslands on north and east facing slopes (Stout et al.
1980). McLean and Wikeem (1985b) found high mortality or severely reduced vigour in
rough fescue plants that were defoliated to a 5 cm stubble height weekly from mid-May
to late June. Willms (1991) also found that rough fescue was very susceptible to negative
impacts of defoliation during the growing season, regardless the frequency or intensity of
clipping. In McLean and Wikeem’s (1985b) study, when clipping was stopped before the
end of vegetative growth or when stubble heights of 10 cm, 15 cm and 20 cm remained,
injury to the plant was reduced. Limiting defoliation to the fall was the only treatment
which did not show adverse effects to, or reduced vigour of, the plant. In a greenhouse
study, King et al. (1998) noted that rough fescue declined in vigour when defoliated
multiple times (clipped three times to 3.5 cm at four week intervals followed by ten
weeks of 16 hour photo-periods) in a season compared to a single defoliation at the end
of the growing season. Summer appears to be the most detrimental time to defoliate
rough fescue because this species appears to have reduced resilience at temperatures that
are above their optimal growing condition of 17oC day time temperature and13oC night
time temperature (King et al. 1998).
Though bluebunch wheatgrass and rough fescue are important forage species and
provide the majority of grazing biomass within their own ecosystems, these species are
mainly located in the Bunchgrass zone of the province; this zone in total makes up
approximately 300,000 hectares (Campbell and Bawtree 1998). When including the
grassland phase of the Interior Douglas-fir zone, grassland area increases to 1.2 million
hectares of the total rangeland within the province (Campbell and Bawtree 1998).
Pinegrass, though providing lower amounts of biomass per unit area than fescue and
bluebunch, is the dominant species in the Interior Douglas-fir and Montane Spruce zones
and is also present in 4 other zones (Ministry of Forests 1997). Within the Interior
Douglas-fir zone alone, pinegrass accounts for 50% of forage and 80% of total plant
cover in the understory (Mclean 1967, McLean 1972).

15

It appears that pinegrass growth periods are related more heavily to environmental
conditions than to specific plant characteristics. This can make it difficult to manage this
species at a specific grazing level without considering environmental conditions such as
precipitation and temperature (Stout and Quinton 1986). The quality of pinegrass forage
decreases significantly as the plant matures, and often is unpalatable to livestock by late
summer (McLean 1967) though some fall re-growth can occur in late August and
September if environmental factors (mainly soil moisture) are favourable (Stout and
Brooke 1985a, Stout et al. 1980). Freyman (1970) conducted clipping research on plots
of pinegrass for three years with twelve different treatments. All clipping was done to a
15 cm stubble height because pinegrass is seldom grazed lower because of unpalatable
material. Plots clipped multiple times during the growing season showed reduced protein
levels versus plots clipped once or twice in August (Freyman 1970). Plots clipped once in
June or once in July and once in August tended to yield more protein than plots clipped
only once or twice in August (Freyman 1970). All clipping treatments (done bi-weekly)
from Stout et al. (1980) resulted in decreased yield in the following year, with clipping to
15 cm during growing season and clipping to 5 cm during dormancy, resulting in the least
amount of yield decrease. Tiller number was also severely affected in clipping treatments
that had a stubble height of 5 cm, and clipping from May 15 to September 15, July 1 to
September 15, and May 15 to August 1. Pinegrass seems to respond negatively to more
severe treatments (higher intensity of biomass removal) but this can vary based on the
time of clipping and environmental factors.
Many studies conclude that grazing pinegrass is most detrimental in mid-July,
after the active growing phase and before summer dormancy (Freyman 1970, Stout et al.
1980, Stout and Brooke 1985a). Freyman (1970) recommended either grazing pinegrass
from July until September, or grazing for two weeks in the beginning of June and again
for two weeks at the beginning of August. Stout et al. (1980) preferred the later
recommendation made by Freyman (1970), but recommended that if the range must be
grazed in July, it should have a subsequent year of rest.

16

There is no shortage of literature on simulated grazing through clipping; however,
a majority of the literature relates the amount of use in terms of stubble height rather than
intensity of clipping. While stubble height gives a clear indication of amount remaining,
percent utilization cannot be determined from this measurement without knowing the
initial size of the plant, or by determining which stubble height equals a specific amount
of photosynthetic material removed. Using specific intensity percentages (e.g. 40%
removal of biomass) and frequency may provide managers with more information on the
impact of grazing by livestock on rangelands.
Another useful source of information on the recovery of range plants after grazing
is initial regrowth. Often in the current literature, clipped plants are measured bi-weekly
or monthly after clipping, or even annually to determine impact on plant vigour. These
measurements are useful and provide important information on plant recovery, but this
does not capture the initial recovery of plants, or the initial impact on plant vigour. Large
changes occur within a plant in the hours after defoliation (inhibited root function and
reduced carbohydrate pools) while certain compensatory processes begin (allocation of
carbon and nitrogen) to aid in the recovery of photosynthetic material (Richards 1993).
Studying the recovery process after defoliation could improve understanding of the
effects environmental factors have on plant regrowth as well as the state of the defoliated
plant during recovery; delayed recovery may result in lower vigour for the remainder of
the growing season (Richards 1993).
The objective of this study was to determine how varying levels of frequency and
intensity of simulated grazing (clipping) impact the vigour of bluebunch wheatgrass,
rough fescue and pinegrass plants as indicated by regrowth, biomass, tiller number and
canopy height. This was done over two growing seasons and looked to meet the
following objectives:
1. Determine if regrowth of bluebunch wheatgrass, rough fescue and pinegrass is
impacted by intensity of treatment or soil moisture within the first two weeks after
clipping.

17

2. Determine how biomass, tiller number and canopy height of bluebunch
wheatgrass, rough fescue and pinegrass is impacted through treatments of various
frequency and intensity clippings and whether intensity and frequency have
interacting effects in biomass, tiller number and canopy height.

2.2 MATERIAL AND METHODS

Study sites
The study was conducted in the Lac du Bois Grasslands Protected area (upper and
lower grasslands) and Pass Lake (Interior Douglas-fir zone), North of Kamloops, British
Columbia. The soils in Lac du Bois are predominantly chernozemic with orthic brown
chernozems in the lower grasslands, orthic black chernozems in the upper grasslands and
orthic luvisols in the Interior Douglas-fir zone (Ministry of Environment, Lands and
Parks 2000). The geographical features in Lac du Bois and surrounding area are primarily
due to past glaciation and glacial deposits (Ministry of Environment, Lands and Parks
2000). The area rises in step-formation from the valley bottom into the higher elevations,
with rolling hills and small mountains in the grassland area reaching up to 1400 m
(Ministry of Environment, Lands and Parks 2000).
This elevation gradient contributes to the presence of a large number of diverse
ecosystems and plant communities, which within the grassland area, are divided into
lower (300-700 m), middle (700-1000 m) and upper grasslands (900-1000 m) (Campbell
and Bawtree 1998). The upper grasslands border the Interior Douglas-fir zone (IDF)
(Ministry of Environment, Lands and Parks 2000). Within these communities there is an
abundance of grass and herbaceous species, some of which dominate more than others. In
the lower grasslands, bluebunch wheatgrass is the dominant grass species and as the
elevation increases into the middle and upper grasslands, rough fescue becomes more
abundant and bluebunch becomes slightly less abundant. Transitioning from open
grasslands into the IDF, bluebunch wheatgrass and rough fescue decrease in abundance

18

and pinegrass becomes the dominant grass species as canopies become more enclosed.
Data collection sites were located near weather stations currently established in Lac du
Bois and in pre-existing exclosures (Fig. 2). These exclosures were built to keep out
cattle but would not limit grazing by other animals.

Figure 2: Aerial image of study sites. Bluebunch wheatgrass and rough fescue sites fall within the Lac du
Bois Grasslands Protected area boundaries, while the pinegrass site lies within an unused pasture on
Agriculture and Agri-Food Canada property (Google Earth 2013).

19

The bluebunch wheatgrass exclosure was located in the lower grasslands of the
Thompson Very Dry Hot Bunchgrass variant (BGxh2) at an elevation of 566 m with a
south facing aspect. The selected research plants within the exclosure were estimated to
occupy 900 m2 of the exclosure while the exclosure itself was approximately 1800 m2.
This area typically receives between 230 mm-250 mm of precipitation annually
(Campbell and Bawtree 1998). Other native plants species found within this area
included big sage brush (Artemisia tridentata), balsam root (Balsamorhiza sagittata),
Sandberg’s bluegrass (Poa secunda), prickly pear cactus (Opuntia fragilis), pasture sage
(Artemisia frigida), long-leaved daisy (Erigeron corymbosus), Thompson’s paintbrush
(Castilleja thompsonii), and sage brush mariposa lily (Calochortus macrocarpus).
The rough fescue exclosure was located in the Thompson Very Dry Hot Interior
Douglas-fir (IDFhx2) variant at an elevation of 903 m with an east facing aspect.
Estimated size of exclosure was determined to be 800 m2, with the study plans occupying
half of this space. Annual precipitation averages vary between 300 mm and 700 mm
(GCC 2011). Rough fescue was the primary plant within the area, but there was a
presence of yellow rattle (Rhinanthus minor), few-flowered shooting star (Dodecatheon
pulchellum), common harebell (Campanula rotundifolia), pasture sedge (Carex
petasata), nodding onion (Allium cernuum), yellow bell (Fritillaria pudica), chocolate
lily (Fritillaria lanceolata), and sticky geranium (Geranium viscosissimum). Prickly rose
(Rosa acicularis) was also found, but only in the north-eastern corner of the exclosure.
The surrounding area contains a few trembling aspen (Populus tremuloides) copses and
isolated Douglas-fir (Pseudotsuga menziesii var. glauca) trees.
The pinegrass exclosure was near Pass Lake in the Thompson Dry Cool Interior
Douglas-fir (IDFdk1) variant at an elevation of 998 m. The area within the exclosure
occupied by the selected plants was approximately 7000 m2, this is such a large number
because plants were spread throughout the exclosure in order to make sure each sample
was located in similar growing conditions to all others. The total area of the exclosure
was approximately 120,000 m2. Annual precipitation averages vary between 300 mm and
700 mm (GCC 2011). The dominant tree species is the Douglas-fir, intermixed with

20

trembling aspen and lodgepole pine (Pinus contorta). A wide array of plant species are
found in this area including shrubs such as soopolallie (Shepherdia canadensis), tall
Oregon-grape (Mahonia aquifolium), common snowberry (Symphoricarpos albus),
twinflower (Linnaea borealis), kinnikinnick (Arctostaphylos uva-ursi) and prickly rose
(Rosa acicularis). In addition, many forbs such as field pussytoes (Antennaria neglecta),
common red paintbrush (Castellija miniata), common harebell (Campanula rotundifolia),
wild strawberry (Fragaria virginiana), northern bedstraw (Galium boreale), tiger lily
(Lilium columbianum), and showy daisy (Erigeron speciosus var. speciosus) are also
found in the area.

Experimental design
For each of the three species, 75 individual plants were selected in a semi-random
fashion. In order to reduce variability, all 75 plants were selected to be of average size
and shape. Plants were marked with pin flags labeled with treatment details and
individually numbered with metal tags staked into the ground near the base of the plant.
Plants were randomly assigned a treatment using a random number generator. Including a
control treatment, a total of five clipping treatments (n=15) were applied:
•

Control – no clipping

•

Clipped once at 40% removal – frequency of 1, intensity of 40% (F1I40)

•

Clipped once at 70% removal – frequency of 1, intensity of 70% (F1I70)

•

Clipped three times at 40% removal – frequency of 3, intensity of 40% (F3I40)

•

Clipped three times at 70% removal – frequency of 3, intensity of 70% (F3I70)

Plants were clipped to approximately forty and seventy percent removal of
photosynthetic material (intensity) and were clipped one or three times, with the multiple
clippings each occurring one week apart (frequency). For multiple frequency clippings,
the second and third removal of biomass was done at 40% of the original volume, not by
remaining volume (all three clippings where done to the same height). All three plant

21

species received the same treatments however, clipping dates varied due to the
elevation/temperature gradient (lower elevations initiate growth before higher elevations)
and clipping also took into consideration when these species are typically grazed on
rangelands in BC. The amount of photosynthetic material removed was determined by
clipping 10 plants from each species one week before initial clippings. Using these
clippings, a height versus weight regression analysis was conducted to determine what
stubble height plants needed to be clipped at to obtain a specific intensity (Appendix 1);
this followed the methods of Lommasson and Jensen (1943). To obtain 40 and 70%
removal, bluebunch wheatgrass was clipped to respective stubble heights of 16.5 cm and
8.3 cm. Rough fescue was clipped to respective stubble heights of 17 cm and 8.7 cm and
pinegrass was clipped to respective stubble heights of 16.4 cm and 8 cm.
Measurements
Clipping and biomass
Timing of clipping was based on growth of plants: boot/head stage for bluebunch
wheatgrass, heading stage for rough fescue, and 3 leaf stage for pinegrass. Both years of
treatments were applied during the same growth stage. In the first year (2011) bluebunch
wheatgrass was clipped initially on May 19th, and twice more at one week intervals for
high frequency treatments. Rough fescue was clipped initially on June 6th, and twice
more at one week intervals for high frequency treatments. Pinegrass was clipped initially
on June14th, and twice more at one week intervals for high frequency treatments. Initial
clippings were bagged, dried and weighed but second and third clippings were not. The
same plants were clipped for the second year of treatment (2012) but initial dates changed
due to growth stage. Bluebunch wheat grass was clipped on May 10th, rough fescue on
May 31st and pinegrass on June 5th. All clippings from each time period were oven dried
in brown paper bags at 60°C until a constant weight was reached and samples were then
weighed. After final measurements in the last year of study, all plants were clipped to
ground level for final biomass.

22

Regrowth
Regrowth was measured on all single clipping treatments (F1I40 and F1I70).
Immediately after initial clipping, ten tillers were randomly selected along a linear line
through each plant, marked with different color paper clips, and were measured by length
to the nearest millimeter. Paper clip color ensured that tiller growth over time was
accurately recorded and tracked. Length measurements were taken again on each of the
five, seven, nine, eleven and thirteen days after initial clipping to see a change in
regrowth over time. The change in length between measurement days was calculated to
achieve cumulative regrowth over 13 days.

Tiller count and canopy height
Live tillers were counted to determine total tiller count. Canopy height was
determined by measuring the ten tallest live tillers and then averaging those numbers. A
tiller was determined to be live as long as it was visibly green. These measurements were
done on all 225 individuals. All samples were measured once prior to treatment
application and then were measured once every two weeks until the beginning of August.
During August, species growth slows due to high temperatures and low soil moisture and
measurements were then taken monthly until the end of the growing season. In 2011, the
final measurements dates were October 6th for bluebunch wheatgrass, October 31st for
rough fescue, and October 4th for pinegrass. In 2012 the last measurement dates for
bluebunch, fescue and pinegrass were September 27th, October 18th, and September 25th
respectively.
Due to high numbers of tillers on rough fescue plants, tillers were only counted
for a quarter of each plant. Measurement triangles were fabricated with a 90 degree angle
with sides of 35 cm and an area of 306 cm2 to ensure the measurement of a quarter of the
plant. Rough fescue plants were measured to find the plant center where a pin flag was
then placed and an additional pin was placed in the right corner of the triangle to ensure
placement of the triangle was consistent. Canopy height for rough fescue was still
measured over the entire plant. Due to the rhizomatous growth form of pinegrass,

23

identifying an individual plant was not a possibility; therefore a 35 cm x 35 cm portable
square was used to designate the measurement area to approximate an individual plant
and obtain an adequate number of tillers for measurements. To mark plots, pin flags were
placed in opposite corners of square to ensure consistency of measurements.

Soil moisture and soil temperature
A POGO Portable Soil Sensor paired with Stevens Hydra Probe II and Stevens
HydraMon Windows CE program was used for determining soil moisture and soil
temperature at each field site. On every measurement day, this probe was inserted to a
depth of 6 cm within 15 cm of the base of each individual plant. The measurements of
soil moisture are given in Volumetric Water Content (VWC) which is essentially a water
to soil ratio. Data are presented in decimal form but can be related to the percentage of
water found in the soil by multiplying by 100.

Statistical analysis
All data were transformed into logarithmic scale for statistical analysis but figures
presented show untransformed data. Biomass, tiller count and canopy height data, even
after transformation, could not be considered normal but the ANOVA is considered
robust enough to accommodate this (McDonald 2009). The Shapiro-Wilk test was used to
determine normality. Non-cumulative regrowth and soil moisture were analyzed using
regressions to determine correlations between the two factors. Cumulative regrowth was
also analyzed for differences between treatments using a linear mixed model (repeated
measures). One-way ANOVA was used to analyze final and cumulative 2012 biomass,
tiller counts and canopy height. A source of variation table for the one-way ANOVA is
shown below (Table 1). Two-way ANOVAs were also used for these measurements to
observe interactions between frequency and intensity. A source of variation table for the
two-way ANOVA is also shown below (Table 2). If ANOVAs revealed a significant
difference (P≤0.05), a Tukey’s HSD post hoc was conducted. Since the tiller counts are a
repeated measures experiment, it was decided to select the highest and lowest tiller
24

counts in each year, for each plant, to create the maximum and minimum tillers counts to
analyze so that standard ANOVAs could be used in place of repeated measures statistics.
Data was only analyzed within each measurement year; there was no statistical
comparison between years. Significance levels were set at P<0.05, however due to
potential ecological significance and importance of this information for managers, Pvalues significant to 0.1 will be discussed as trends. All data was analyzed using the
statistical program R (R Core Team 2012).

Table 1: Source of variation for one-way ANOVA.

Source

Degrees of
Freedom

Mean Square

F

Treatment ( between groups)

5-1=4

SSt/df

MSt/MSe

Error ( within groups)

75-5=70

SSe/df

Total

75-1=74

Table 2: Source of variation for two-way ANOVA.

Source

Degrees of
Freedom

Mean Square

F

2-1=1

SSa/dfa

MSa/MSe

Factor A (Frequency)
(between groups)
Factor B (Intensity)
(between groups)
Interaction

2-1=1

SSb/dfb

MSb/MSe

(2-1)(2-1)=1

SSab/dfab

MSab/MSe

Error (within groups)

60-4=56

SSe/dfe

Total

60-1=59

25

2.3 RESULTS
Weather
Weather stations near each species site have been in place since 1979 and long
term averages were used to create growing season climate normal’s for each site over a
period of 32 years (1979-2010) with one exception of pinegrass climate normal’s in April
which reflected 8 years (2002-2010). Data from the 2011 and 2012 measurement years
have been compiled together with climate averages in figures 3-5. The bluebunch
wheatgrass site appears to have had both average high and average low temperatures in
2011 and 2012 that were outside of the climate normal (Fig. 3a and 3b). Precipitation in
2011 was slightly above normal in May, but appeared to be lower in the fall. Precipitation
in 2012 peaked in June significantly above the climate normal (Fig. 3c). The rough fescue
site had temperature ranges within the standard deviation of the climate normal in both
years (Fig. 4a and 4b) while precipitation showed the same trend as for the bluebunch
wheatgrass site (Fig. 4c). The pinegrass site had temperatures within the standard
deviation of the climate normal in both years (Fig. 5a and 5b). Again precipitation in both
years is greater in the spring and lower in the fall when compared to the climate normal
(Fig. 5c).

26

Temperature (°C)

a. 40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0

Normal
2011
2012

Month

Temperature (°C)

b. 16.0
12.0
8.0
4.0
0.0

Normal

-4.0

2011

-8.0

2012

-12.0

c.

120.0

Precipitation (mm)

Month

100.0
80.0
60.0
Normal

40.0

2011

20.0

2012

0.0

Month
Figure 3: Climate normal and 2011 and 2012 weather data for bluebunch wheatgrass site. a). Average
high temperature (°C). b). Average low temperature (°C). c). Precipitation (mm). Error bars on climate
line indicate standard deviation of untransformed data.

27

a.

30.0

Temperature (°C)

25.0
20.0
15.0
Normal

10.0

2011

5.0

2012

0.0

Month

Temperature (°C)

b.

14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
-2.0
-4.0

Normal
2011
2012

Month

Precipitation (mm)

c. 120.0
100.0
80.0
60.0
Normal

40.0

2011

20.0

2012

0.0

Month
Figure 4: Climate normal and 2011 and 2012 weather data for rough fescue site. a). Average high
temperature (°C). b). Average low temperature (°C). c). Precipitation (mm). Error bars on climate line
indicate standard deviation of untransformed data.

28

a.

30.0

Temperature (°C)

25.0
20.0
15.0
Normal

10.0

2011

5.0

2012

0.0

Month

Temperature (°C)

b.

12.0
10.0
8.0
6.0
4.0
2.0
0.0
-2.0
-4.0
-6.0
-8.0

Normal
2011
2012

Month

Precipitation (mm)

c. 120.0
100.0
80.0

n

60.0
40.0
20.0

Normal
2011
2012

0.0

Month
Figure 5: Climate normal and 2011 and 2012 weather data for pinegrass site. a). Average high
temperature (°C). b). Average low temperature (°C). c). Precipitation (mm). Error bars on climate line
indicate standard deviation of untransformed data. * April climate average only includes 8 years of
data.

29

Regrowth and soil moisture
Bluebunch wheatgrass
In 2011, the statistical analysis of cumulative regrowth between the low intensity
treatment (I40) and high intensity treatment (I70) showed no significance (Fig. 6a) (P≤
0.05), (DF = 144, P = 0.0878). Using the soil moisture data, a regression was done for
2011 non-cumulative regrowth and 2011 soil moisture collected at the time of regrowth
measurements. While the R-squared value was relatively low (R2 = 0.1111) the
regression showed a correlation between non-cumulative regrowth and soil moisture (F =
16.21, DF = 131, Significance F = 9.55E-05). In 2012, cumulative regrowth of the low
intensity and high intensity treatments was similar (Fig. 6b) and no significance between
treatments was found (DF = 144, P = 0.11). The regression of soil moisture and noncumulative regrowth for 2012 showed a slightly higher R-squared value than that in 2011
(R2 = 0.2737) and a correlation between non-cumulative regrowth and soil moisture (F =
23.83, DF = 68, Significance F = 6.82E-06).

4

0.35

3.5

0.3

3

0.25

2.5

0.2

2
0.15

1.5
1

Intensity 40

0.5

Soil moisture

Intensity 70

0.1

Soil moisture (VWC)

Regrowth (cm)

a.

0.05

0

0
5

7

9
Day

11

13

30

b.
8

0.35
Intensity 70

Regrowth (cm)

7

0.3

Soil moisture

6

0.25

5

0.2

4
0.15

3

0.1

2

Soil moisture (VWC)

Intensity 40

0.05

1
0

0
5

7

9
Day

11

13

Figure 6: Cumulative regrowth (cm) and soil moisture (VWC) of bluebunch wheatgrass, 5, 7, 9, 11, and
13 days after removing 40% or 70% of total biomass. a.) Bluebunch wheatgrass regrowth and soil
moisture in 2011. b.) Bluebunch wheatgrass regrowth and soil moisture in 2012. Data in graph represents
untransformed data and error bars are for untransformed data while statistical analysis was done with
transformed data.

Rough fescue
In 2011, cumulative regrowth of the low intensity treatment was not different than
that of the high intensity treatment (Fig. 7a) (DF = 144, P = 0.1248). The regression of
soil moisture and non-cumulative regrowth for 2011 showed a very low R-squared value
(R2 = 0.0242) and showed no correlation between regrowth and soil moisture (F = .442,
DF = 141, Significance F = 0.507). In 2012, cumulative regrowth of the low intensity and
high intensity treatments was similar (Fig. 7b) and no significance between treatments
was found (DF = 144, P = 0.3261). The regression of soil moisture and non-cumulative
regrowth for 2012 showed a slightly higher R-squared value than that in 2011 (R2 =
0.0799) and correlation between non-cumulative regrowth and soil moisture was found (F
= 12.54, DF = 147, Significance F = 0.000534).

31

8

0.35

7

0.3

Regrowth (cm)

6

0.25

5

0.2

4
0.15

3
2
1

Intensity 40

0.1

Intensity 70

0.05

Soil moisture

0
5

7

9
Day

11

Soil moisture (VWC)

a.

0

13

8

0.35

7

0.3

Regrowth (cm)

6

0.25

5

0.2

4
0.15

3
2

Intensity 40

1

Intensity 70
Soil moisture

0
5

7

9
Day

11

0.1

Soil moisture (VWC)

b.

0.05
0

13

Figure 7: Cumulative regrowth (cm) and soil moisture (VWC) of rough fescue, 5, 7, 9, 11, and 13 days
after removing 40% or 70% of total biomass. a.) Regrowth and soil moisture of rough fescue in 2011. b.)
Regrowth and soil moisture of rough fescue in 2012. Data in graph represents untransformed data and error
bars are for untransformed data while statistical analysis was done with transformed data.

32

Pinegrass
In 2011, cumulative regrowth of the low intensity treatment was lower than the
high intensity treatment on each of the measurement days (Fig. 8a); statistical analysis
showed significance (DF = 149, P = 0.0072). The regression of soil moisture and noncumulative regrowth for 2011 showed a low R-squared value (R2 = 0.0806) and there was
no significant correlation between non-cumulative regrowth and soil moisture (F = 3.697,
DF = 140, Significance F = 0.0565). In 2012, cumulative regrowth of the low intensity
and high intensity treatments was very similar (Fig. 8b) and no significance between
treatments was found (DF = 149, P = 0.4376). The regression of soil moisture and noncumulative regrowth for 2012 found a slightly higher R-squared value than that in 2011
(R2 = 0.1104) while correlation between non-cumulative regrowth and soil moisture was
found (F = 17.76, DF = 148, Significance F = 4.36E-05).

4.5

0.4

4

0.35

3.5

0.3

3

0.25

2.5

0.2

2

0.15

1.5

Intensity 40

0.1

1

Intensity 70

0.5

Soil moisture

0

Soil moisture (VWC)

Regrowth (cm)

a.

0.05
0

5

7

9
Day

11

13

33

4.5

0.4

4

0.35

3.5

0.3

3

0.25

2.5

0.2

2

0.15

1.5

Intensity 40

1

Intensity 70

0.5

Soil moisture

0

0.1

Soil moisture (VWC)

Regrowth (cm)

b.

0.05
0

5

7

9
Day

11

13

Figure 8: Cumulative regrowth (cm) and soil moisture (VWC) of pinegrass, 5, 7, 9, 11, and 13 days after
removing 40% or 70% of total biomass. a.) Pinegrass regrowth and soil moisture for 2011. b.) Pinegrass
regrowth and soil moisture for 2012. Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data.

Final biomass
Final biomass represents the product of complete clipping of the above ground
material for each plant at the end of the 2 year study. No interactions for final biomass
were observed for bluebunch wheatgrass or pinegrass. Rough fescue did show a
significant interaction term between frequency and intensity (F = 5.149, Residuals = 56,
P = 0.0272) (Fig. 9).
Bluebunch wheatgrass final biomass in treatments with low intensity (F1I40,
F3I40) did not differ significantly from one another, regardless of frequency. However,
clipping at a high intensity (F1I70, F3I70) significantly reduced final biomass regardless
of whether the clipping was once (F1) or three times (F3). This data indicates that
intensity is having more of an impact on bluebunch wheatgrass final biomass than
frequency (Fig. 10a).

34

Rough fescue did not show as clear of an effect for higher intensity of clipping yet
high intensity at high frequency (F3I70) still showed statistical significance in lower final
biomass when compared to other treatments (Fig. 10b).
The results for pinegrass differed from both bluebunch and rough fescue.
Pinegrass final biomass shows a declining trend as treatment becomes more severe (Fig.
10c). All treatments were significantly different from the most severe treatment, while the
least severe treatment was also significantly different from the F3I40 treatment.

Mean final biomass dry weight (g)

40
35
30

Intensity 40
A

Intensity 70

A

A

25
20

B

15
10
5
0

Frequency 1
Frequency 3
Figure 9: Interactions between frequency and intensity for
rough fescue final biomass (g). Data in graph represents
untransformed data and error bars are for untransformed data
while statistical analysis was done with transformed data.

35

Cumulative 2012 biomass
Cumulative biomass from 2012 included the final biomass (complete clipping of
the plants at the end of the season) plus all of the biomass removed during each of the
clipping events.
Bluebunch wheatgrass displayed a significant difference in biomass occurring in
plants clipped at higher intensity when compared to lower intensity treatments (Fig. 10d).
The control treatment was significantly different from both high intensity treatments but
not significantly different from the low intensity treatments.
Rough fescue showed no significant difference in cumulative biomass across most
of the treatments. The most severe treatment of high frequency/high intensity (F3I70) had
a significantly lower cumulative biomass than all of the other treatments (Fig. 10e).
In pinegrass, the control was not significantly different from any of the
treatments. However the F1I70 treatment had higher final biomass than both of the high
frequency treatments (Fig. 10f). The control and F1I40 treatment do display a trend of
being higher than both high frequency treatments (P<0.1).
No interactions were found for bluebunch wheatgrass or pinegrass, but there was
an interaction found between intensity and frequency for rough fescue (F = 10.053,
Residuals = 56, P = 0.0025) (Fig. 11).

36

a. 14

14
A

10
A

8
6

A

B

B

4

AB

10
BC

8

C

6
4

2

2
0

0
F1I40

F1I70

F3I40

Control F1I40 F1I70 F3I40 F3I70

F3I70

Bluebunch wheatgrass

Bluebunch wheatgrass

e. 60

b. 60

50
A

A
A

30

B
20

Dry weight (g)

Dry weight (g)

50
40

20

0
Control F1I40 F1I70 F3I40 F3I70

F3I70

Rough Fescue

Rough Fescue

f. 7

6

6

5

5
A
AB

3

B
C

2

Dry weight (g)

Dry weight (g)

7

4

A

30

0
F3I40

A

B

10

F1I70

A
A

40

10

F1I40

c.

A

12
Dry weight (g)

Dry weight (g)

12

d.

AB

AB

A

B

B

4
3
2
1

1
0

0
F1I40

F1I70

F3I40

Pinegrass

F3I70

Control F1I40 F1I70 F3I40

F3I70

Pinegrass

Figure 10: Final biomass (g) and cumulative biomass 2012 (g) clippings from 2012 of all three plant species.
a.) Bluebunch wheatgrass final biomass. b.) Rough fescue final biomass. c.) Pinegrass final biomass. d.)
Bluebunch wheatgrass cumulative biomass. e.) Rough fescue cumulative biomass. f.) Pinegrass cumulative
biomass. Letters denote a significance level of P<0.05. Data in graph represents untransformed data and error
bars are for untransformed data while statistical analysis was done with transformed data. Control – No
clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped once at 70% removal, F3I40 – clipped three
times at 40% removal, F3I70 – clipped three times at 70% removal.

37

Intensity 40
Intensity 70

Mean cumulative 2012 biomass dry
weight (g)

60
A
50

A

A

40
B
30
20
10
0

Frequency 1
Frequency 3
Figure 11: Interactions between frequency and intensity for
rough fescue cumulative biomass 2012 (g). Data in graph
represents untransformed data and error bars are for
untransformed data while statistical analysis was done with
transformed data.

Maximum tillers
In 2011 maximum tiller counts for all species showed no significant differences
between treatments (Appendix 2, 3 and 4). In the second year of measurements, there
were significant differences shown in two of the three species at varying levels. No
interactions between frequency and intensity were found for maximum tiller count in any
species in both 2011 and 2012. For bluebunch wheatgrass, multiple clippings at a high
intensity (F3I70) resulted in decreased tiller numbers when compared to all treatments
except the F1I70 (Fig. 12).

38

Maximum tiller count

180
160
140
120
100
80
60
40
20
0

A

A

A
AB
B

Control F1I40

F1I70

F3I40

F3I70

Bluebunch Wheatgrass 2012
Figure 12: Maximum tillers counts of bluebunch wheatgrass plants in
2012. Letters denote significance levels (P<0.05). Data in graph
represents untransformed data and error bars are for untransformed
data while statistical analysis was done with transformed data. Control
– No clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped
once at 70% removal, F3I40 – clipped three times at 40% removal,
F3I70 – clipped three times at 70% removal.

Rough fescue maximum tiller counts resulted in no significant difference among
any of the treatments in either 2011 or 2012.
For pinegrass there was no treatment effect in 2011 but in 2012 the multiple
frequency (F3) treatments resulted in a lower number of tillers than the F1I70 treatment
but neither were different from the control or F1I40 treatment (Fig. 13).

39

160

Maximum tiller count

120

A

A

140
AB

AB

100

B

80
60
40
20
0
Control F1I40

F1I70

F3I40

F3I70

Pinegrass 2012
Figure 13: Maximum tillers counts of pinegrass plants in 2012.
Letters denote significance levels (P<0.05). Data in graph represents
untransformed data and error bars are for untransformed data while
statistical analysis was done with transformed data. Control – No
clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped once
at 70% removal, F3I40 – clipped three times at 40% removal, F3I70
– clipped three times at 70% removal.

Minimum tillers
There were no significant interactions between frequency and intensity for any
species in either 2011 or 2012 for minimum tiller count. For bluebunch wheatgrass the
F3I70 treatment resulted in a significantly lower number of tillers than all of the other
treatments in 2011 (Fig. 14a). In 2012 both the F1I70 and F3I70 treatments had lower
tiller numbers (Fig. 14b). This is a similar trend to biomass and maximum tiller count for
this species.
Minimum tiller number was not significant by treatment for rough fescue in 2011
and 2012 or for pinegrass in 2011(Appendix 5 and 6). Pinegrass minimum tiller counts in
2012 show that the control treatment is not significantly different from any of the other
treatments, however, the most severe treatment (F3I70) has a significantly lower number
of tillers than the remaining clipping treatments (Fig. 15).

40

Minimum tiller count

b.

180
160
140
120
100
80
60
40
20
0

A
A

A

A
B

A

180
160
140
120
100
80
60
40
20
0

A

A
B

B

Control F1I40 F1I70 F3I40 F3I70

Control F1I40 F1I70 F3I40 F3I70

Bluebunch Wheatgrass 2011

Bluebunch Wheatgrass 2012

Figure 14: Minimum tiller counts for bluebunch wheatgrass. a.) Minimum tiller counts for 2011. b.)
Minimum tiller counts for 2012. Letters denote significance levels (P<0.05). Data in graph represents
untransformed data and error bars are for untransformed data while statistical analysis was done with
transformed data. Control – No clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped once
at 70% removal, F3I40 – clipped three times at 40% removal, F3I70 – clipped three times at 70%
removal.

40
35
Minimum tiller count

Minimum tiller count

a.

30
A

25
20

A
A

AB

15
B

10
5
0
Control

F1I40

F1I70

F3I40

F3I70

Pinegrass 2012
Figure 15: Minimum tiller counts for pinegrass in 2012. Letters denote significance
levels (P<0.05). Data in graph represents untransformed data and error bars are for
untransformed data while statistical analysis was done with transformed data. Control
– No clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped once at 70%
removal, F3I40 – clipped three times at 40% removal, F3I70 – clipped three times at
70% removal.

41

Canopy height
All species over both years showed varying degrees of difference for canopy
height among treatments. In 2011, bluebunch wheatgrass showed a gradual decrease in
canopy height as severity of the treatment increased (Fig. 16a). In 2012 the control
canopy height was significantly different from all other treatments while those of both
low intensity treatments were significantly different from the heights of high intensity
treatments (Fig. 16b). The low frequency/high intensity treatment also had significantly
different canopy height from the most severe treatment, which was significantly different
from all treatments.

b.
70

70
60
50
40

A
B
C

C

30

D

20
10
0
Control F1I40 F1I70 F3I40 F3I70
Bluebunch Wheatgrass 2011

Maximum canopy height (cm)

Maximum canopy height (cm)

a.

60
50
40
30

A
B

B
C

D

20
10
0
Control F1I40 F1I70 F3I40 F3I70
Bluebunch Wheatgrass 2012

Figure 16: Bluebunch wheatgrass canopy height (cm) for 2011 and 2012 seasons. a.) 2011 canopy
height. b.) 2012 canopy height. Letters denote significance levels (P<0.05). Data in graph represents
untransformed data and error bars are for untransformed data while statistical analysis was done with
transformed data. Control – No clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped once
at 70% removal, F3I40 – clipped three times at 40% removal, F3I70 – clipped three times at 70%
removal.

42

Rough fescue in both 2011 and 2012 showed significantly higher canopy for the
control from all clipping treatments. The most severe treatment (F3I70) had a
significantly lower canopy height from all clipping treatments. All of the other treatments
were not significantly different from one another (Fig. 17). Significant interactions
between frequency and intensity were also found for rough fescue canopy heights in 2011
(F = 6.644, Residuals = 56, P = 0.01266) and in 2012 (F = 6.476, Residuals = 56, P =
0.01377) (Fig. 18) but not for bluebunch wheatgrass or pinegrass.

100
90
80
70
60
50
40
30
20
10
0

b.
A

B

B

B

C

Maximum canopy height (cm)

Maximum canopy height (cm)

a.

100
90
80
70
60
50
40
30
20
10
0

A
B

B

B
C

Control F1I40 F1I70 F3I40 F3I70

Control F1I40 F1I70 F3I40 F3I70

Rough Fescue 2011

Rough Fescue 2012

Figure 17: Rough fescue canopy height (cm) for 2011 and 2012 seasons. a.) 2011 canopy height. b.) 2012
canopy height. Letters denote significance levels (P<0.05). Data in graph represents untransformed data
and error bars are for untransformed data while statistical analysis was done with transformed data. Control
– No clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped once at 70% removal, F3I40 –
clipped three times at 40% removal, F3I70 – clipped three times at 70% removal.

43

Mean of maximum canopy height
(cm)

Intensity 40

b.

Intensity 70
52

A

A

A

47
B
42

37

32

Mean of maximum canopy height
(cm)

57

a.

57

Intensity 40
Intensity 70

52

47

A

A

A

42
B
37

32
Frequency 1

Frequency 3

Frequency 1

Frequency 3

Figure 18: Interactions between frequency and intensity of clipping in rough fescue
canopy height (cm) in 2011 and 2012. a.) Rough fescue canopy height 2011 interactions.
b.) Rough fescue canopy height 2012 interactions. Data in graph represents untransformed
data and error bars are for untransformed data while statistical analysis was done with
transformed data.

The pinegrass control treatment had a higher canopy measurement than all other
treatments in both 2011 and 2012. For each year, the result of the clipping treatments on
canopy height was slightly different (Fig. 19) but the general trend was a reduction in
height with increasing intensity and frequency of clipping. There was little to no
significant difference between F1I40, F1I70 and F3I40 treatments in both years. Control
was always significantly different from all other treatments.

44

a. 60

b. 60

A

40

A

50
BC

B
CD

D

30
20
10
0
Control F1I40 F1I70 F3I40 F3I70
Pinegrass 2011

Maximum canopy height (cm)

Maximum canopy height (cm)

50

B

40

BC

BC

C

30
20
10
0
Control F1I40 F1I70 F3I40 F3I70
Pinegrass 2012

Figure 19: Pinegrass canopy height (cm) for 2011 and 2012 seasons. a.) 2011 canopy height. b.) 2012
canopy height. Letters denote significance levels (P<0.05). Data in graph represents untransformed data
and error bars are for untransformed data while statistical analysis was done with transformed data.
Control – No clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped once at 70% removal,
F3I40 – clipped three times at 40% removal, F3I70 – clipped three times at 70% removal.

2.4 DISCUSSION
Regrowth and soil moisture
Regrowth varied between species and between years, though it appears that in
2012 regrowth between treatments was very similar within each species. Soil moisture
was found to be correlated with regrowth most years for the three species. In cases where
the correlation was not significant to P=0.05, it was shown to be significant to P=0.1.
Due to the soil probe being limited to a depth of 6 cm, it represents a coarse estimate of
soil moisture available to the plants. There is little literature on these three species that
relates specifically to initial regrowth of mature plants.

45

Bluebunch wheatgrass
While there was no statistically significant (P<0.05) results for regrowth of
bluebunch wheatgrass in 2011 or 2012 there was a strong trend (P<0.1) in 2011 which
indicated the results may have ecological significance and should be discussed. It has
been shown that the canopies of bluebunch wheatgrass plants direct and concentrate rain
water to the base of the plant (Ndawula-Senyimba et al. 1971) and the loss of aerial plant
parts may result in reduced soil moisture recharge through decreased snow capture and
increased frost damage to plants (Wikeem et al. 1989). The low intensity clipping would
leave more canopy than the higher intensity treatment allowing for greater collection of
rainwater at the plant’s base; this appears to have occurred in 2011 when soil moisture
was at a much higher level than in 2012. The water collection mechanism would provide
greater opportunity for regrowth for plants with larger canopy size. With the low soil
moisture levels in 2012, the canopy size of a plant after treatment most likely had little
effect on opportunity for rainfall direction and concentration as little to no rainfall was
recorded during the regrowth measurement period. This likely resulted in the same
growing opportunity which is shown by almost identical regrowth patterns between
treatments in 2012. No winter data is available to determine if there was more snow cover
and less frost damage on the lower intensity plants however the low soil moisture levels
in 2012 may be partially attributed to reduced soil moisture recharge. While regrowth is
not explained by soil moisture in either 2011 or 2012, there is a strong correlation with
regrowth and soil moisture indicating that increased soil moisture is associated with
increased regrowth. In 2011, soil moisture was likely not limiting, but was found to be
correlated with regrowth. In 2012 soil moisture appeared to be limiting, and this
limitation is likely correlated with the low regrowth of both treatments in 2012.

Rough fescue
The lack of a significant difference in rough fescue regrowth across treatments in
both years could be related to the interactions between frequency and intensity found in

46

some of the rough fescue vigour parameters. A single clipping at the prescribed
intensities was likely not severe enough to cause significant difference in regrowth
response between treatments. In 2011 soil moisture generally had a decreasing trend
while regrowth continued to increases at a steady rate, though no correlation was found in
this year to P=.0.05, there is likely an ecological significance to this as this correlation
would be significant at P=0.1 and soil moisture likely did play some role in regrowth for
rough fescue in this year. In both years soil moisture did not appear to be limiting nor
over adequate. The exclosure in which rough fescue was studied was a long term
exclosure which results in the plants likely having very high vigour after many years of
not being grazed by livestock.

Pinegrass
In 2011, the high intensity treatment (I70) regrew significantly more than the low
intensity treatment (I40), contrary to the expected outcome. This trend could be explained
by either the high intensity clipping allowing more sunlight through (Grant et al. 1981) to
lower, newer leaves, or by resource sharing. Allowing more sunlight through the canopy
may have allowed for lower, previously shaded, leaves to become more photosyntheticlly
active (Grant et al. 1981). Soil temperature was relatively constant between treatments
(averaging 20°C during regrowth measurements), so extra sunlight did likely not increase
soil temperatures in high intensity treatments to encourage more growth. Stout and
Brooke (1985b) found that rhizome connections in pinegrass between cut and uncut
tillers allowed for the exchange of food reserves. Other rhizomatous grasses such as
tussock grass (Schizachyrium scoparium) and big bluestem (Andropogon gerardi) can
share resources between tillers (Derner 2012, Archer and Delting 1984). Plants
surrounding the pinegrass plots were not clipped during the trial therefore it is possible
that tillers outside of plots were sharing resources with the cut tillers within plots. It is
possible that the high intensity clipping created a “resource sink” where treated plants
drew in more resources from surrounding tillers than the low intensity treatment, which
cut to a height of 17cm, may not have triggered the movement of incoming resources. In

47

2012 over all regrowth was slightly greater than 2011 and soil moisture was found to
impact regrowth in this year; soil moisture in 2012 was recorded to be greater than that in
2011 particularly during last 4 measurement dates. Soil moisture did not appear to be
limiting in either years during the regrowth measurement period, nor was it over
adequate. No correlation was found between soil moisture and regrowth in 2011 though
there is a strong trend (P<0.1) suggesting that there still might be some relation between
the two factors, though there are many other factors influencing regrowth as well. Soil
moisture and regrowth correlation in 2012 again shows that while regrowth is not
explained solely by soil moisture, it is an influencing factor.

Measurements of plant vigour
Final and cumulative 2012 biomass, tiller counts and canopy heights were all used
to evaluate plant vigour response to clipping treatments. These measurements were found
to be some of the most common measurements used in published research papers to
quantify how various treatments affect plant vigour. Counting reproductive tillers is
another measurement method often used, however it was not used in this study because
all three plant species needed to have the same measurements recorded and because
pinegrass is an infrequently-flowering grass species (Stout and Quinton 1986). For the
purposes of this discussion, the vigour of a plant is related to the general size of the plant
and its parts (leaves and tillers), growth rate of the plant and its ability to reach a larger
size within a plant population (Price 1991).

Bluebunch wheatgrass
A consistent trend among the vigour parameters for bluebunch wheatgrass was
that higher intensity clipping had a greater effect on plant vigour than frequency. This
trend was significant in all measurements except for canopy height in 2011. Intensity may
impact bluebunch wheatgrass vigour more than frequency for several different reasons.
An early study of bluebunch wheatgrass by Wilson et al. (1966) noted that different

48

frequencies at various times of the growing season resulted in different implications for
plant growth. It was found that clipping in the boot stage resulted in the highest impact on
plant vigour likely due to greater effects on root growth and carbohydrate accumulation
during this stage (Wilson et al. 1966). Clipping earlier in the season would allow new
leaves to grow which could meet the photosynthetic needs of the plant allowing for
continued root growth and carbohydrate storage (Wilson et al. 1966). In our experiment,
bluebunch wheatgrass was cut two years in a row during boot stage which could explain
why intensity impacted plant vigour to a greater degree than frequency. The high
intensity treatment (when compared to the lower intensity treatment) may have impacted
carbohydrate reserves, removed more elevated growing points and removed more
photosynthetic material which would cause a decrease in photosynthesis occurring within
a plant. McLean and Wikeem (1985a) determined that leaving more foliage resulted in
less damage to the plant and suggested that this was due to a better ability to replenish
carbohydrate reserves as a result of the photosynthetic material remaining. McLean and
Wikeem also cited Wilson et al. (1966) observing that a 20 cm clipping height, compared
to clipping more intensively to lower stubble heights, improved plant yields. Removal of
growing points may also result in the death of tillers which leads to production of new
tillers; this is a much slower process when compared to leaf expansion of surviving tillers
(Dahl and Hyder 1977). The high intensity treatments probably saw a net loss in tiller
number due to less new tillers being produced than were lost.

Rough fescue
Consistent results across rough fescue vigour parameters were also found with the
most severe treatment (F3I70) always being significantly different from control and other
clipping treatments. Rough fescue was also the only species to show interaction effects
for final and cumulative biomass, and canopy height in both 2011 and 2012. Final
biomass and canopy height interactions are additive, where the combined effects of
frequency and intensity impact the plant response. Cumulative 2012 biomass interactions
show a reversed interaction (though not significant) where high frequency treatments

49

show a reversed interaction compared to the low frequency treatments. Some literature
suggests that frequency has a greater effect on plant vigour than intensity however it is
stated that rough fescue is quite susceptible to any type of defoliation during the growing
season (King et al. 1998, McLean and Wikeem 1985b, Willms and Fraser 1992, Willms
1991). Based on the literature it was expected that all treatments would be significantly
different from the control in terms of biomass however this was not the case. The
clipping treatments were conducted when rough fescue plants were leaving the boot stage
and entering into the heading. At this time rough fescue can be sensitive to defoliation of
growing point locations. The similarities between treatments for biomass results may be
linked to the amount of photosynthetic material remaining after clipping. McLean and
Wikeem (1985b) found stubble heights of 10, 15 and 20 cm resulted in reduced injury to
plants compared to a stubble height of only 5 cm. For this experiment the low intensity
clipping (40% intensity) of rough fescue yielded a stubble height of 17 cm while the high
intensity clipping (70% intensity) yielded a stubble height of 8.7 cm. The remaining
stubble height on both intensity treatments may have been sufficient for the treated plants
to maintain a high enough level of photosynthesis to mitigate clipping impacts. In the
case of the most severe treatment, F3I70, the combination of a high frequency and high
intensity was enough to impact plant growth and subsequently, biomass.
The lack of significance in all rough fescue tiller measurements is not
representative of what has been found in other studies. Willms and Fraser (1992) found
that rough fescue tiller numbers were impacted by intensity and frequency of clipping.
Plants being cut at lower frequencies had similar tiller numbers to that of an unclipped
control and a one-time clipping event and appeared to stimulate tiller growth possibly
because of removal of self-shading. The lack of significance in our data could potentially
be related to the methodology used to count tillers; if green was visible on a tiller, it was
considered live. From personal observation it was noted that typically 2 or 3 leaves grow
out from the crown of one tiller. Often what was seen on the higher severity treatments
(e.g. F3I70) was that the majority of leaves on one tiller were dead, while one green leaf
remained and so this tiller was counted as live. This resulted in the more severe

50

treatments having a larger amount of dead rather than live material compared to control
and less severe treatments (e.g. F1I40) even though when tillers were counted, the
numbers were approximately the same.

Pinegrass
Pinegrass did not have as clear of a trend as bluebunch wheatgrass or rough
fescue but typically the high frequency and high intensity treatment (F3I70) showed a
consistently, significantly lower vigour response than most other treatments. Stout and
Quinton (1986) demonstrated that clipping biweekly for successive years to stubble
heights of 5 cm or 10 cm decreased stand density and that when stubble heights of 15 cm
were used a 20% decrease was found. It is important to note that these differences were
not found to be significant after 4 years (Stout and Quinton 1986). Freyman (1970)
indicated that pinegrass yield was higher in treatments with low clipping frequencies
which is similar to the results found in the experiment presented in this thesis. The
cumulative 2012 biomass did show that the higher frequency treatments had lower yields
than the F1I70 treatment, but were not significantly different from control and F1I40.
However due to a trend in the control and F1I40 having elevated yields when compared
to the high frequency treatments (P<0.1) these findings still agree with those of Freyman
(1970).
Maximum and minimum tiller counts were only significant in 2012 but the results
showed that the control was not significantly different from any of the treatments; there
were only significant differences between some of the treatments. In maximum tiller
counts, both low frequencies were significantly different from the high frequency high
intensity clipping and there was a trend that the low frequency treatments had slightly
greater tiller counts than the F3I40 treatment (P<0.1). This trend was not found in
minimum tiller counts for 2012. These results could again be related to the findings of
Freyman (1970) with frequency impacting pinegrass growth greater than intensity.
The variability of significance between treatments for the various vigour
measurements could be more closely related to environmental conditions or grazing

51

history of the plant; Stout et al. (1980) stated that these two factors influence a plants
response to defoliation and to observe a distinct plant response to treatments, multiple
years of manipulation may be required. While the F3I70 treatment is consistently the
lowest mean of measurements groups, it is often not significantly different from the
control means for cumulative 2012 biomass and minimum and maximum tiller counts in
2012.

2.5 CONCLUSION
All three species responded to treatments in different ways which is expected as
each species has its own unique physiological and morphological characteristics and ideal
growing conditions. Though the responses are dissimilar, this data does highlight the idea
that management of rangelands should keep frequency and intensity of grazing under
careful consideration. The response of these species to the treatments might change based
on season of use and competition from other plants. This study only clipped selected
species and did not clip surrounding vegetation. In a grazing scenario, other plants would
be grazed as well resulting in different competitive effects as Mueggler (1972) showed
that reduced competition can offset the effects of clipping.
In British Columbia, the focus of management tends to be on stocking rates and
stubble heights which in themselves are indicators of intensity; however, grazing
rotations are very common in the Southern Interior which often results in livestock being
in pastures for a long duration. This duration of animals in a pasture can typically mean a
high frequency of defoliation for plants. For a species like bluebunch wheatgrass,
managing only for intensity may make sense as this appears to be what affects plant
vigour the most, but for other species such as rough fescue (interactions between
frequency and intensity) and pinegrass (high variability) incorporating other factors such
as frequency into management considerations would be beneficial. Opportunity for
growth or regrowth should always be considered when making management decisions as
precipitation (soil moisture), temperature and disturbances can all impact plant vigour.

52

CHAPTER 3: EVALUATING THE GRAZING RESPONSE INDEX

3.1 INTRODUCTION
Chapter 2 presented results of a clipping study that manipulated frequency and
intensity of clipping on three different plant species that dominate plant communities in
the Southern Interior of British Columbia. This study was designed in a manner that
manipulated frequency and intensity at specific levels in order to determine how these
factors influenced plant vigour. The results and discussion presented in chapter 2 will aid
in the determination of the answers to the following management question: Can the
Grazing Response Index be used to predict plant response to grazing in the Southern
Interior?

3.2 THE GRAZING RESPONSE INDEX
In the 1990s, Colorado State University Range Extension Program and Integrated
Resource Management Program developed a tool to evaluate the effects of grazing in a
current year while integrating climatic factors and thus the growing conditions of forage
species (Reed et al. 1999). The information collected from this tool allows land managers
to gain beneficial use of available forage resources while maintaining a vigorous plant
community. This tool has been found to be effective and simple by the Rocky Mountain
Region Forest Service who has adopted this approach for evaluating grazing impacts
(Reed et al. 1999). The grazing response index (GRI) centers around three basic factors
relating to plant health:


Frequency – The number of times a plant is subject to defoliation.



Intensity – The amount of photosynthetic material that is removed at one time of
defoliation.



Opportunity – The opportunity a plant has to grow before defoliation, or regrow
after defoliation. This is linked closely with environmental conditions.

53

Frequency
This factor is important for plant recovery after defoliation as the more times a
plant is defoliated, the more energy it must put into regrowing new tissue to account for
the lost photosynthetic capacity. Studies by Briske (1986) determined that seven to ten
days is needed for a plant to initiate enough regrowth for it to be selected for grazing
another time. At this time the plant has invested enough energy (in the form of
carbohydrates) into the new tissue that another defoliation event may have negative
implications. These studies also showed that if defoliation occurred three or more times
in one growing season, plant vigour was negatively impacted. If this trend continued,
plants would have a lessened ability to persist in the community (Ellison 1960). Table 3
shows how to score the frequency of defoliation. To determine how many times a plant
will be subject to defoliation, the number of days livestock are in a pasture (or a known
portion of pasture) is divided by the regrowth period of seven to ten days. If plant
regrowth is faster, seven days is used to determine frequency, while if plant growth is
slower, ten days is used. Seven days is considered to be a conservative number if exact
growth rate is not known (weather can also play a factor in regrowth) (Reed et al. 1999).
For example, if livestock are in a pasture for 14 days then individual plants were exposed
to potential defoliation at least twice (14/7 = 2) and the GRI score would be 0.

Intensity
The most important part of intensity of defoliation (percent utilization of the
plant) is the total amount of photosynthetic material that remains on the plant, as this is
what will enable the plant to recover from the event (Reed et al. 1999). Intensity of
clipping is also important in terms of the removal of growing points. This is dependent
upon the growth stage of the plant, but generally, higher intensity defoliations may
remove more growing points than lower intensity defoliations. Table 3 shows the scoring
of intensity. Intensity is visually estimated immediately after animals are removed from

54

pasture. The use of cages, placed in representative areas, can be an effective way to help
determine intensity levels (Reed et al. 1999).

Opportunity
Opportunity for the plant to grow before defoliation, or re-grow after defoliation
will have the most impact on plant health and vigour compared to the other two factors.
This is why opportunity has a larger scoring scale (Reed et al. 1999). Table 3 shows the
scoring for opportunity. Scoring opportunity is based on an individual’s judgment of the
appearance of the vegetation at the end of the growing season. The score is influenced by
environmental factors such as precipitation and temperature as well as various
disturbances such as fire, insects, and excess trampling. It also takes into account the
growth stages of plants when cut, as removal of growing points at stages of growth such
as boot stage or flowering stage can be more detrimental to a plant than at a vegetative
stage.

Table 3: The GRI scoring values for frequency, intensity and opportunity. Adapted from Reed et al. 1999.

Frequency
(Number of defoliations)
1 (≤ 7 days)
2 (8 - 14 days)
3 or more (>14 days)
Intensity
(Amount of biomass removed)
Light (≤ 40 %)
Moderate (41-55%)
Heavy (≥ 56 %)
Opportunity
(To grow before grazing or regrow
after grazing)
Full Season
Most of Season
Some Chance
Little Chance
No Chance

Value
1
0
-1
Value
1
0
-1
Value
2
1
0
-1
-2

55

Creating a score
After scoring out each portion of the GRI, numbers are totaled and the resulting
number (positive, neutral or negative) will provide an evaluation of the grazing
management. A positive score means the plant community is benefiting from the
management, neutral score means the management is neither harmful nor beneficial to
the community, and a negative score means the management is harming the community
and overall health and vigour is decreasing (Reed et al. 1999). One year with a negative
score may not cause significant damage but continuing to manage grazing at the same
levels will result in continued degradation.
The GRI is directly linked to the management of a rangeland, where stocking
rates, timing of grazing, and duration of grazing play an important role in each of the
three index measurements (Reed et al. 1999). The GRI should be scored on a yearly basis
which will allow land managers to see changes in plant vigour after each season. In larger
pastures, multiple sites may need to be scored as some areas may receive heavier use than
others; areas where livestock access water will often receive heavier use than other areas
which are a greater distance from water sources.
The GRI allows for adjustments to management and planning between
assessments of current plant community and desired plant community. Essentially, the
GRI provides opportunity for short term management in conjunction with long term
management and long term rangeland assessments. Referring back to chapter 1, the range
condition and range health assessments are conducted at intervals of several years and
have a greater depth of assessment than the GRI, however; applying the GRI to
rangelands on an annual basis would allow for more frequent evaluation of vegetation
conditions. The accumulation of annual scores can provide the opportunity to observe
potential trends (positive or negative) of the plant community’s response to grazing. By
observing these scoring trends, managers can make the decision to change or maintain
management plans at the end of every grazing season while using the current assessments
every few years to confirm the accuracy of trends being observed through the GRI.

56

3.3 EVALUATING THE GRAZING RESPONSE INDEX BASED ON PLANT
VIGOUR RESPONSE TO CLIPPING TREATMENTS
To address the management objective “Can the Grazing Response Index be used
to predict plant response in the Southern Interior?” the differences in plant response
should be addressed. Using the GRI in the Southern Interior would result in the scoring of
many different plant communities with the same indices and if the differences are too
distinct, then the scores given by the GRI may not accurately reflect the response of all
plant communities in this region.
The outcome of results for bluebunch wheatgrass (summarized in Appendix 7a)
indicates that intensity impacts plant vigour more than frequency (Fig. 10a,d). This
becomes more apparent in the second year of treatments for tiller counts (Fig. 12 and
14a,b) and canopy height (Fig. 16a,b) as significance levels change to show lower
intensity treatments general have more tillers or higher canopies than the high intensity
treatments.
While the tiller counts of rough fescue were unfortunately not reflective of what
was likely happening with regards to plant vigour (Appendix 3 and 5), final and
cumulative biomass (Fig. 10b,e) are good indicators of plant vigour and canopy height
measurements (Fig. 17a,b) also showed significant differences (summarized in Appendix
7b). The rough fescue results present a pattern that indicates the most severe treatment is
always significantly different from the control and all other clipping treatments. We did
see an additive effect of intensity and frequency in the interactions for both biomass
measurements as well as for canopy height in both years. This indicates that defoliation at
only high frequencies or only high intensity may not be enough to impact plant vigour,
however high frequency and high intensity clipping will impact plant vigour.
Similar to rough fescue, pinegrass typically saw the most severe treatment (F3I70)
being significantly different from most other treatments, though there were no
interactions found. Several vigour measurements (cumulative 2012 biomass, minimum
and maximum tiller counts in 2012) found that control was often not significantly
different from the most severe treatment to P<0.05 however a general trend was that

57

control exhibited slightly higher means than the F3I70 treatment (P<0.1) (Appendix 7c).
Overall, the results of pinegrass are quite variable compared to that of rough fescue and
bluebunch wheatgrass. This variability may be related to the grown form of pinegrass
(rhizomatous) or environmental conditions.
With this review of results it becomes clear that these three forage species do not
respond to defoliation in a similar manner. There were differences in response to
treatments and these differences are most likely due to many factors such as plant
physiology and growth form (bunchgrass vs. rhizomatous grass) as well as weather and
resources. These plants were also chosen representatives of their plant communities as
they are the dominant forage species, but their responses to these treatments may not
reflect responses of the rest of the plant community. Though each plant species showed
different responses in measurement to the treatments, the overall indications lead me to
conclude that the general trend is decreasing vigour for increasing severity of treatment.
The reason the GRI has the potential to be such a useful tool, is because it is
simplified. To adjust the tool for each individual plant community based on slightly
different responses would cause the GRI to lose its effective methodology. The GRI was
created with the intention of being used over a wide variety of plant communities and its
flexibility and simplicity in scoring leads to the conclusion that it should be robust
enough to score the plant communities in the Southern Interior.

Scoring of three key forage species by treatment
I scored the treatments (F1I40, F1I70, F3I40 and F3I70) using the GRI methods
and compared these scores with general plant vigour response to treatment based on the
results of biomass, tiller counts and canopy height. Scoring was done for the last year of
treatments (2012) as this is when more significant differences were found within vigour
parameters and when the final biomass measurements were obtained. Final biomass is an
important measure as it has been shown to be the plant component most affected by
defoliation (Ferraro and Oesterheld 2002).

58

Frequency scoring
For frequency it is deemed that seven to ten days are required for enough
regrowth to occur that the removal of this regrowth through another grazing event would
impact plant vigour. Seven days is used as a conservative number if plant growth is fast,
and ten days if plant growth is slower. Scoring this will require knowledge about the
dominant plant species in the pasture being scored and how they grow.
For bluebunch wheatgrass regrowth in 2012 the low intensity (40%) and the high
intensity (70%) treatments had an average of 1.9 and 1.8 cm of regrowth respectively
(Fig. 6b). Removing this amount of regrowth would likely impact the vigour of
bluebunch wheatgrass plants as they were cut during boot/head stage when carbohydrate
reserves are typically low (Miller et al. 1986) and soil moisture is limiting on sites
bluebunch wheatgrass typically grows on, therefore seven days is appropriate to score
bluebunch wheatgrass for frequency of grazing.
Regrowth for rough fescue in 2012 after seven days was averaged at 2.1 and 2.3
cm for low intensity and high intensity treatments respectively. Removing this amount of
regrowth would likely impact rough fescue vigour and it has been shown that frequency
is often a factor of grazing that will negatively impact rough fescue vigour, and grazing
rough fescue while it is actively growing will likely result in some decrease of vigour
(King et al. 1998, McLean and Wikeem 1985b, Willms and Fraser 1992, Willms 1991).
Seven days is an appropriate number to use to score frequency for rough fescue.
Pinegrass results for regrowth indicated an average of 1.7 cm for both intensity
treatments. While 1.7 cm may not be a significant amount of regrowth, pinegrass tends
to have higher yields when cut at lower frequencies (Freyman 1970) and the fact the
environmental conditions play an important role in pinegrass growth (Stout et al. 1980)
it is best to use the more conservative application of seven days for frequency.
With all three species determined to be impacted by a second grazing event after
seven days, frequency scores for each treatment (for each species) are shown in table 4.

59

Intensity scoring
Intensity scoring for the GRI is based on the amount of photosynthetic material
(by biomass, not by height) removed at the end of the grazing period. While this is based
on the amount removed, what is important about intensity is the amount of photosynthetic
material remaining after grazing that will enable the plant to recover. Bluebunch
wheatgrass clipping intensity results indicated that lower intensity treated plants had
significantly more biomass (final and cumulative) and higher tiller counts and canopy
heights and generally were not significantly different from the control plants which lead
to the conclusion that a removal of 40% intensity did not negatively impact plant vigour.
The high intensity treatments were significantly lower than control which indicates a
negative impact on plant vigour.
Due to the interactions observed with rough fescue, there was no clear indicator of
how these plants responded to the intensity of treatment, however, when a high intensity
was paired with high frequency for a treatment, there was always a significant difference
between this treatment and all others. Initial vigour of these plants would likely mitigate
response to high intensity of treatment (as all rough fescue plants in this study were
contained in a long term exclosure) however, if this study was conducted for several
more years, trends in intensity may have become more apparent over time. Taking this
into consideration, the categories of intensity presented by the GRI would likely reflect
rough fescues response to intensity of defoliation.
Pinegrass results were variable and make it difficult to give a clear suggestion as
to which intensity categories each treatment correlates with. Intensity might be easier to
categorize if pinegrass was clipped at a later time, but for this study it was at the three
leaf stage (vegetative growth) and therefore growing points were likely not affected
resulting intensity not having as great of an impact as frequency. Grazing does not always
take place during vegetative growth of pinegrass; therefore it is better to remain with
slightly more conservative scoring.
All three species were determined to fit into the categories of intensity presented
in the GRI and thus the scoring for intensity for each treatment is shown in table 4.

60

Opportunity scoring
With clipping dates and weather data, it would be possible to determine a relative
opportunity score, but this component of the GRI is meant to be scored visually at the end
of the growing season over an area based on appearance of the pasture. This is difficult to
do with plants dispersed around an exclosure and without there being bias between the
treatments as the results of impact of plant vigour have already been shown and
discussed.
Opportunity is often the most difficult aspect of the GRI to score as there are
many considerations to make. Climate conditions, timing of defoliation, past use and
disturbance can all play a role in a plant opportunity to grow or regrow which is why the
opportunity carries the most weight in producing the final GRI score. Bluebunch
wheatgrass was assigned slightly different opportunity for growth and regrowth for two
main reasons: time of clipping and weather data indicated that precipitation in 2012 was
outside of the standard deviation of the climate average at time of clippings (May) and
average high temperature in 2012 was outside the standard deviation of climate normal
throughout all of 2012. Temperature and precipitation for rough fescue and pinegrass
sites where not outside of the standard deviation of the climate normal.
All control plants receive an opportunity score of +2 (full season) as they were not
clipped. The F1I40 treatments received a score of +1 (most of season) for all three
species as this treatment was not significantly different from the control in any of the
vigour parameter measurements (with the exception of canopy heights). The F1I70
treatment for bluebunch wheatgrass opportunity received a score of 0 (some chance),
where rough fescue and pinegrass received a score of +1. The opportunity score of
bluebunch wheatgrass was decreased by one for this treatment because of timing of
clipping which likely resulted in the removal of growing points as bluebunch wheatgrass
was in boot stage at this time. The initiation of new tillers likely results from these
clippings, which can be a slow process therefore the opportunity for regrowth was
diminished. Scoring for the F3I40 treatment opportunity was 0 for all species because of
multiple clippings in the beginning/middle of the growing season resulted in the plants

61

have less time to regrow compared to single frequency clippings. The most severe
treatment, F3I70 was scored as a 0 for rough fescue and pinegrass. There was some
change for growth and regrowth for these species. This treatment was scored as a -1 for
bluebunch wheatgrass, again because of the high intensity removing growing points and
the multiple clippings. Visual observations of most plants receiving this treatment
showed insignificant amount of regrowth. All opportunity scores and total scoring for
each species is shown in table 4.

Table 4: The GRI scores for each treatment based on 2012 results.

Treatment

Frequency
score
All
Species

Intensity
score
All
Species

Opportunity:
Bluebunch
wheatgrass

Opportunity:
Rough fescue
and pinegrass

Total Score:
Bluebunch
wheatgrass

Total Score:
Rough
fescue and
pinegrass

Control

+1

+1

+2

+2

+4

+4

F1I40

+1

+1

+1

+1

+3

+3

F1I70

+1

-1

0

+1

0

+1

F3I40

-1

+1

0

0

0

0

F3I70

-1

-1

-1

0

-3

-2

Evaluation conclusion
It is important to keep in mind that the growing seasons for these species are not
all the same which may result in different opportunity scores by species. An elevational
and temperature gradient results in bluebunch wheat grass initiating growth before the
other two species, while rough fescue initiates growth before pinegrass. All species
become dormant when summer temperatures rise above optimal growth levels. In some
cases, bluebunch wheatgrass sites may have lower opportunity because these areas are
more prone to low soil moisture levels and higher temperatures than the rough fescue and
62

pinegrass sites. Over all, these relative opportunity scores do provide useful information
in terms of presenting a completed GRI score, though opportunity was not directly
manipulated or researched within this experiment aside from collecting soil moisture and
weather data.
After reviewing the scoring of the GRI by each category for each species, the
final scores were produced. These scores provide an accurate indication of plant response
to these defoliation treatments. Comparing the statistical analysis to the scores given by
the GRI, it is realized that there may be some discrepancies. An example of such
discrepancy would be the pinegrass control plants not being significantly different from
even the most severe treatments, which is not reflected in the scoring of the GRI where
the control plants receive a +4 and the most severe treatment (F3I70) receives a -2.
Firstly, this study was conducted over a period of two growing season; a relatively short
time to obtain definitive answers. Secondly, it is likely that the scoring of the GRI might
be conservative on certain treatments. Perhaps the most severe treatment for pinegrass
did not respond as negatively to treatments a predicted by the GRI, however over several
more years, repeating this treatment would likely result in negative consequences for
these plants. This indicates a conservative scoring of the most severe treatment for
pinegrass; better to be cautious and score conservatively for such treatments as it is a
strong possibility that this treatment would become more detrimental over time.
Conservative scoring of other treatments may be occurring as well. The bluebunch
wheatgrass treatment F3I40 being scored the same as the F1I70 treatment (final score of
0), even though the F3I40 treatment showed significantly higher biomass means. While
the F3I40 may have show higher means in the plant vigour measurements, perhaps an
extension of this study would reveal an eventual change in significance. The GRI is
showing a trend that indicates the continued defoliation of this kind would not benefit the
vigour of these plants under these conditions. Given a longer application of the GRI to
real grazing occurrences, the scores given would provide an accurate indication of
vegetation conditions over time for these three forage species.

63

3.4 FINAL CONCLUSION
The rangelands of the Southern Interior of British Columbia are diverse and unique
among the rangelands of North America and require diligent and consistent management
to ensure continued sustainable use of these valuable ecosystems. The livestock industry
is heavily reliant on the proper functioning of these rangelands and a host of other social
and economical values (watersheds, recreation, wildlife, forestry, aboriginal values) can
be impacted if rangelands become degraded. Currently the range condition and range
health assessments are two long term monitoring tools used in British Columbia for
rangeland management that provide useful and accurate representation of the status of
vegetation and over all ecosystem function. Rangelands in the Southern Interior do have a
history of over-grazing from the early 1900s and because of proper management many of
these systems have recovered or are recovering. It is not a question of if current
management is working effectively, but a question of how can we continue to improve
upon what we already know. To strive for continued improvement of our management
strategies is what will guarantee continued sustainable use and decreased degradation on
the range.
Management with a short term assessment tool like the Grazing Response Index
(GRI) in conjunction with the long term assessments will give managers, but also
livestock producers themselves, a tool that can be used every growing season to track
short term changes in vegetation. The GRI breaks down the three main principles of plant
growth (frequency, intensity and opportunity) to a simplistic form that still remains
effective. Using this tool would allow range managers and producers alike to manage for
more than just stocking rates and utilization but for duration in a pasture and even
climatic conditions.
Plant response results from Chapter 2 indicate that the dominant forage species in
the Southern Interior respond in different ways to different treatments of frequency and
intensity but overall, the more severe the treatment, the more significant the decrease in
plant vigour. The use of the GRI by land managers or livestock producers does require

64

knowledge about the rangeland or pastures that are being scored, and this knowledge will
help to determine the differences in response to frequency and intensity so different plant
communities can be scored accurately by managers.
It is realized that this study was short term. To obtain a clearer understanding of
how the GRI will truly work on the rangelands in the Southern Interior, it must be
incorporated into management practices and applied to real grazing events, not just
simulated grazing (clipping). Once used over a period of years, rangeland condition
assessments and rangeland health assessments should provide an accurate account of how
the GRI is predicting plant response to grazing in this region. The implementation of the
GRI in rangeland management in British Columbia will require some practice and
training for managers and livestock producers in order to ensure correct usage and
understanding of the tool and its indices. Once this process of introduction of the GRI to
rangelands in British Columbia is completed, future management can be adjusted on an
annual basis that will best benefit the livestock producers as well as the plant
communities.
To summarize the GRI in simpler terms think of the native rangelands being a
bank account. In good years investments are made into the land to increase its value
(increasing vigour and plant health) but eventually there will be bad years when
withdrawals will need to be made (heavier grazing, drought, disturbance). Bad years will
occur, but if the investments are greater than the withdrawals there will always be a
positive balance (healthy, vigourous plants). With that idea, consider the GRI to be the
rangelands balance book. On a yearly basis scores are given to determine if an investment
(positive score) or a withdrawal (negative score) is made. Tracking these investments and
withdrawals gives a year to year statement on the wealth of your account (vigour of your
plants). Now, because the GRI is not a stand-alone tool, consider the long term rangeland
condition assessments to be your audits. Audits are required every few years to make sure
the balance book isn’t being cheated. The rangeland assessments ensure that the GRI

65

scoring is reflective of the actual plant community and gives a much clearer picture on
the overall vigour of rangeland plant species.

66

LITERATURE CITED
Archer, S., Detling, J.K. 1984. The effects of defoliation and competition on regrowth of
tillers of two North American mixed-grass prairie graminoids. Oikos. 43(3):351357.
Bawtree, A.H. 2005. A history of range use in British Columbia. Rangelands. 27(6):3639.
B.C. Cattlemen’s Association. 2013. B.C. Beef Cattle Industry. Accessed on August 11,
2013. URL: http://www.cattlemen.bc.ca/overview.htm
Blaisdell, J.P., Pechanec, J.F. 1949. Effects of herbage removal at various dates on vigour
of bluebunch wheatgrass and arrowleaf balsamroot.Ecology. 30(3): 298-305.
Borman, M.M., Pyke, D.A. 1994. Successiona theory and the desired plant community
approach. Rangelands. 16:82-84.
Branson, F.A. 1956. Quantitative effects of clipping treatments on five range grasses.
Journal of Range Management. 9(2):86-88.
Briske, D.D. 1986. Plant response to defoliation: morphological considerations and
allocation priorities. pp. 425-427. In: P.J.Joss, P.W. Lynch and O.B. Williams
(eds.) Rangelands: A Resource under Siege. Cambridge University Press.
Briske, D.D., Fuhlendorf, S.D., Smeins, F.E. 2005. Sate-and-Transition models,
thresholds and rangeland health: A synthesis of ecological concepts and
perspectives. Rangeland Ecology and Management. 58:1-10.
Briske, D.D., Derner, J.R., Brown, S.D., Fuhlendorf, W.R., Teague, K.M. 2008.
Rangeland Ecology and Management. 61:3-17.
Caldwell, M.M. 1984. Plant requirements for prudent grazing. Developing strategies for
Rangeland Management. Westview Press, Boulder, Colorado; pp.117-152.
Campbell, C. W., Bawtree, A.H. 1998. Rangeland handbook for BC. Pp. 1-203. B.C.
Cattlemens Association. Kamloops, B.C.
Clark, P.E., Krueger, W.C., Bryant, L.D., Thomas, D.R. 1998. Spring defoliation effects
on bluebunch wheatgrass: II. basal area. Journal of Range Management. 50(5):
526-530.

67

Dahl, B.E., and D.N. Hyder. 1977. Developmental morphology and management
implications. P. 258-290. In: R.E. Sosebee (ed.) Rangeland Plant Physiology.
Range Sci. Series No. 4, Soc. Range Manage., Denver, Colo.
Delesalle, B.P., Coupe, B.J., Wikeem, B.M., Wikeem, S.J. 2009. Grasslands Monitoring
Manual for British Columbia: A Tool for Ranchers. Grasslands Conservation
Council of British Columbia. Accessed on: March 11, 2011. Available from:
http://www.bcgrasslands.org/monitoringmanual.htm
Derner, J.J, Briske, D.D., Polley, W.H. 2012. Tiller organization within the tussock grass
Schizachyrium scoparium: a field assessment of competition-cooperation
tradeoffs. Botany. 90:669-677.
Dyksterhuis, E.J. 1949. Condition and management of rangeland based on quantitative
ecology. Journal of Range Management. 2:104-115.
Ellison L. 1960. Influence of grazing on plant succession of rangelands. Botanical
Review. 26:1-78.
Ferraro, D.O., Oesterheld, M. 2002. Effect of defoliation on grass growth. A quantitative
review. Oikos. 98(1):125-133.
Fraser, D.A. 2009. Range resources assessment procedures. 2nd ed. B.C. Min. For.
Range. Range Br., Kamloops, B.C. Rangeland Health Brochure 9. Accessed on:
March 11, 2011. Available from: http://www.for.gov.bc.ca/hra
Freyman, S. 1970. Effects of clipping on pinegrass.Canadian Journal of Plant Science.
50: 736-739.
Ganskopp, D.C., Bedell, T. 1981. An assessment of vigour and production of range
grasses following drought. Journal of Range Management. 34:137-141.
Google Earth. 2013. Google Earth. Asscessed on July 10, 2013. URL:
http://earth.google.com
Grant, S.A., Barhtram, G.T., Torvell, L. 1981 Components of regrowth in grazed and cut
Loliumperenne swards. Grass and Forage Science. 36:155-168.
Grasslands Conservation Council of British Columbia (GCC). 2011. Where are BC’s
grasslands? Accessed on March 11, 2011. URL:
http://www.bcgrasslands.org/wherearebcgrasslands.htm
Hamann, A. & Wang, T. 2006. Potential effects of climate change on ecosystem andt
species distribution in British Columbia. Ecology. 87:2773–2786.

68

Hilebert, D.W., Swift, D.M., Detling, J.K., Dyer, M.I. 1981. Relative growth rates and
grazing optimization hypothesis. Oecologia. 51:14-18.
Holechek, J.L., Baker, T., Boren, J.C., Galt, D. 2006. Grazing impacts on rangeland
vegetation: What we have learned. Rangelands. 28(1):7-13.
Holechek, J.L., Piper, R.D., Herbel, C.H. 2011. Range management principles and
practices. 6th ed. Prentice Hall, One Lake street, Upper Saddle River, N.J.
Johnson, W.M. 1956. The effect of grazing intensity on plant composition, vigour and
growth of pine-bunchgrass ranges in central Colorado. Ecology. 37:790-798.
King, J.R., Hill, M.J., Willms, W.D. 1998. Temperature effects on regrowth of 3 rough
fescue species. Journal of Range Management. 51(4): 463-468.
Lommasson, T., Jensen, C. 1943. Determining utilization of range grasses by heightweight tables. Journal of Forestry. 41 (8): 589-593.
McDonald, J.H. 2009. Handbook of biological statistics (2nd ed.) Sparky House
publishing, Baltimore, Maryland. Pp150-154. Accessed on: July 7, 2013. URL:
http://udel.edu/~mcdonald/statnormal.html
McLean, A. 1967. Beef production on lodgepole pine-pinegrass range in southern British
Columbia. Journal of Range Management. 20(4): 214-216.
McLean, A. 1972. Beef production on lodgepole pine-pinegrass range in the Cariboo
region of B.C. Journal of Range Management. 25:10-11.
McLean, A., Wikeem, S. 1985a. Season and intensity of defoliation on bluebunch
wheatgrass survival and vigour in southern British Columbia. Journal of Range
Management. 38(1): 21-26.
.
McLean, A., Wikeem, S. 1985b. Rough fescue response to season and intensity of
defoliation. Journal of Range Management. 38(2): 100-103.
Meays, C.L., Laliberte, A.S., Doescher, P.S. 2000. Defoliation response of bluebunch
wheatgrass and crested wheatgrass: Why we cannot graze these two species in the
same manner. Rangelands. 22(6): 16-18
Miller, R.F., Seufert, J.M., Haferkamp, M.R. 1986. The ecology and management of
bluebunch wheatgrass (Agropyron spicatum): A review. Agricultural Experiment
Station. Oregon State Univesity. Station bulletin 669.
Milroy, J.E., McLean, A. 1980. History of range administration and research in British
Columbia. Rangelands. 2(2):56-59.

69

Ministry of Forests. 1997. The Pinegrass complex. Accessed on: January 18, 2013. URL:
http://www.for.gov.bc.ca/hfp/publications/00049/pinegras.pdf.
Ministry of Environment, Lands and Parks. 2000. Lac du Bois grasslands park
management plan background document. Kamloops BC. Accessed on: Janurary
25, 2011. URL:
http://www.env.gov.bc.ca/bcparks/planning/mgmtplns/lacdubois/lacdubois.pdf.
Ministry of Forests, Lands and Natural Resource Operations, 2013a. Range Act.
Accessed on: July 16, 2013. URL:
http://www.bclaws.ca/EPLibraries/bclaws_new/document/ID/freeside/00_04071_
01#part1_division2
Ministry of Forests, Lands and Natural Resource Operations, 2013b. Range Program.
Accessed on July 16, 2013. URL:
http://www.for.gov.bc.ca/hra/Legislation/grazing.htm
Mueggler, W.F. 1972. Influence of competition on the response of bluebunch wheatgrass
to clipping. Journal of Range Management. 25:88-92.
Ndawula-Senyimba, M.S., Brink, V.C., Mclean, A. 1971. Moisture interception as a
factor in the competitive ability of bluebunch wheatgrass. Journal of Range
Management. 24(3): 198-200.
Newman, R., Krizic, Wallace, B. 2011. Linking range health assessment methodology
with science: rough fescue grasslands of British Columbia. B.C. Min. For.
Range, For. Sci. Prog. Victoria, B.C. Tech. Rep. 062. Availavle online at:
www.for.gov.bc.ca/hfd/pubs/Docs/Tr/Tro62.htm
Price, P.W. 1991. The plant vigor hypothesis and herbivore attack. Oikos. 62:244-251.
Pyke, D.A., Herrick, J. E., Shaver, P., Pellant, M. 2002. Rangeland health attributes and
indicators for qualitative assessment. Journal of Range Management. 55(6): 584597.
Quinton, D.A., Mclean, A., Stout, D.G. 1982. Vegetative and reproductive growth of
bluebunch wheatgrass in interior British Columbia. Journal of Range
Management. 35(1): 46-51.
R Core Team. 2012. R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria.
ISBN 3-900051-07-0, URL http://www.R-project.org/.

70

Reed, F., Roath, R., Bradford, D. 1999. The grazing response index: A simple and
effective method to evaluate grazing impacts. Rangelands. 21(4):3-6.
Richards, J.H. 1993. Physiology of plants recovering from defoliation.Proceedings of the
XVII International Grassland Congress. 1993:85-94.
Rustad L.E. 2008. The response of terrestrial ecosystems to global climate change:
Towards an integrated approach. Science of the Total Environment. 404: 222235.
Ryle, G.J.A., Powell, C.E. 1975. Defoliation and regrowth in graminaceous plant: the
role of current assimilate. Annals of Botany. 39:297-310.
Society for Range Management. 1989. A glossary of terms used in range management.
3rd ed. Denver, Colorado.
Statistics Canada. 2013. Cattle Industry by Province. Accessed on August 11, 2013.
URL: http://www.statcan.gc.ca/tables-tableaux/sum-som/l01/cst01/prim50aeng.htm
Stout, D.G., Brooke, B. 1985a. Growth and development of pinegrass in the interior
British Columbia. Journal of Range Management. 38(4): 312-317.
Stout, D.G., Brooke, B. 1985b. Rhizomes and root below clipped pinegrass tillers have a
higher percentage of carbohydrate when attached to other non-clipped tillers.
Journal of Range Management. 38:276-277.
Stout, D.G., McLean, A., Brooke, B., Hall, J. 1980. Influence of simulated grazing
(clipping) on pinegrass growth. Journal of Range Management. 33(4): 286-291.
Stout, D.G., Quinton, D.A. 1986. Pinegrass: an important forage species in Interior B.C.
Research Branch, Agriculture Canada. Kamloops British Columbia.
Vallentine, J.E. 1990. Grazing Management. Acedemic Press Inc., San Diego, California.
Pp398.
Westoby, M., Walker, B., Noy-Meir, I. 1989. Opportunistic management for rangelands
not at equilibrium. Journal of Range Management. 42(4): 266-274.
Wikeem, B.M., Newman, R.F., van Ryswyk, A.L. 1989. Effect of fertilization and litter
removal on grassland forage production. Journal of Range Management.
42(5):412-415.
Wikeem, B.M., McLean, A., Bawtree, A.H., Quinton, D.1993. An overview of forage
resources and beef production on Crown land in British Columbia. Canadian
Journal of Animal Science. 73:779-794.

71

Wikeem, B. And Wikeem, S. 2004. Grasslands of British Columbia.Grassland
Conservation Council of British Columbia.Accessed on: March 21, 2013. URL:
http://www.bcgrasslands.org
Willms, W.D. 1991. Cutting frequency and cutting height effects on rough fescue and
parry oat grass yields. Journal of Range Management. 44(1): 82-86.
Willms, W.D., Fraser, J. 1992. Grwoth characteristics of rough fescue (Fesctuca
scabrella var.campestris) after three years of repeated harvesting at scheduled
frequencies and heights. Canadian Journal of Botany. 70:2125-2129.
Wilson, A.M., Harris, G.A., Gates, D.H. 1966. Cumulative effects of clipping on yield of
bluebunch wheatgrass. Journal of Range Management. 19:90-91.

72

APPENDICES
Appendix 1

a.

100
90
80

Biomass removed (%)

70
60

y = -2E-06x3 + 0.062x2 - 5.1904x + 108.63
R² = 0.9752

50
40
30
20
10
0
0

10

20
30
Stubble height (cm)

40

50

60

120

b.
Biomass removed (%)

100
80
60

y = 0.0606x2 - 5.16x + 110.13
R² = 0.9666

40
20
0
0

10

20

30

40

50

60

Stubble height (cm)

Biomass removed (%)

c.

120
100
80
60
y = -0.0008x3 + 0.1274x2 - 6.8824x + 121.26
R² = 0.9618

40
20
0
0

10

20

30

40

50

60

70

Stubble height (cm)
Appendix 1: Regressions of stubble height (cm) versus plant biomass (g) for each species. a.)
Bluebunch wheatgrass. b.) Rough fescue. c.) Pinegrass. Untransformed data used for regression.

73

Appendix 2

180
Maximum tiller count

160

A

A

A

F1I40

F1I70

A

A

F3I40

F3I70

140
120
100
80
60
40
20
0
Control

Bluebunch Wheatgrass 2011
Appendix 2: Maximum tiller counts for bluebunch wheatgrass in 2011. Letters denote significance
(P<0.05). Data in graph represents untransformed data and error bars are for untransformed data while
statistical analysis was done with transformed data. Control – No clipping, F1I40 – clipped once at 40%
removal, F1I70 – clipped once at 70% removal, F3I40 – clipped three times at 40% removal, F3I70 –
clipped three times at 70% removal.

74

Appendix 3

b.

180
160
140
120
100
80
60
40
20
0

A

A

A

A

A

Control F1I40 F1I70 F3I40 F3I70
Rough Fescue 2011

Maximum tiller count

Maximum tiller count

a.

180
160
140
120
100
80
60
40
20
0

A
A

A

A

A

Control F1I40 F1I70 F3I40 F3I70
Rough Fescue 2012

Appendix 3: Maximum tiller counts for rough fescue. a.) Rough fescue maximum tiller count in
2011. b.) Rough fescue maximum tiller counts in 2012. Letters denote significance (P<0.05). Data
in graph represents untransformed data and error bars are for untransformed data while statistical
analysis was done with transformed data. Control – No clipping, F1I40 – clipped once at 40%
removal, F1I70 – clipped once at 70% removal, F3I40 – clipped three times at 40% removal, F3I70
– clipped three times at 70% removal.

75

Appendix 4

160

Maximum tiller count

140

A

A
A

A

A

120
100
80
60
40
20
0
Control

F1I40

F1I70

F3I40

F3I70

Pinegrass 2011
Appendix 4: Maximum tiller counts for pinegrass in 2011. Letters denote significance (P<0.05). Data in
graph represents untransformed data and error bars are for untransformed data while statistical analysis was
done with transformed data. Control – No clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped
once at 70% removal, F3I40 – clipped three times at 40% removal, F3I70 – clipped three times at 70%
removal.

76

Appendix 5

b.

a.

140

Minimum tiller count

140

80

A
A

A
A
A

40

Minimum tiller count

100

60

A

120

120

100

A
A

A

A

80
60
40

20

20

0

0
Control F1I40 F1I70 F3I40 F3I70

Control F1I40 F1I70 F3I40 F3I70

Rough Fescue 2011

Rough Fescue 2012

Appendix 5: Minimum tiller counts for rough fescue. a.) Rough fescue minimum tiller count in 2011.
b.) Rough fescue minimum tiller counts in 2012. Letters denote significance (P<0.05). Data in graph
represents untransformed data and error bars are for untransformed data while statistical analysis was
done with transformed data. Control – No clipping, F1I40 – clipped once at 40% removal, F1I70 –
clipped once at 70% removal, F3I40 – clipped three times at 40% removal, F3I70 – clipped three
times at 70% removal.

77

Appendix 6
45

Minimum tiller count

40

A

A
A

A

35
A

30
25
20
15
10
5
0
Control

F1I40

F1I70

F3I40

F3I70

Pinegrass 2011
Appendix 6: Minimum tiller counts for pinegrass in 2011. Letters denote significance (P<0.05). Data in
graph represents untransformed data and error bars are for untransformed data while statistical analysis was
done with transformed data. Control – No clipping, F1I40 – clipped once at 40% removal, F1I70 – clipped
once at 70% removal, F3I40 – clipped three times at 40% removal, F3I70 – clipped three times at 70%
removal.

78

Appendix 7
Bluebunch wheatgrass

a.

Final
Biomass
(g)

Cumululative
2012 (g)

Max.
Tiller
2011

Max.
Tiller
2012

Min.
Tiller
2011

Min.
Tiller
2012

Canopy
Height
2011(cm)

Canopy
Height
2012(cm)
Si
μ
g.

Treatment

μ

Sig.

μ

Sig.

μ

Sig.

μ

Sig.

μ

Sig.

μ

Sig.

μ

Sig.

Control

N/A

N/A

11.6

A

131

A

153

A

58

A

153

A

57.5

A

42.5

A

F1I40

9.2

A

11

A

138

A

157

A

33

A

106

A

39.4

B

35.3

B

F1I70

3.9

B

7.1

BC

140

A

125

AB

34

A

67

B

32.1

C

27.9

C

F3I40

6.8

A

8.6

AB

151

A

157

A

31

A

105

A

31.5

C

33.8

B

F3I70

2.9

B

5.6

C

146

A

108

B

16

B

49

B

27.5

D

25.2

D

Rough fescue

b.

Min.
Tiller
2011

Min.
Tiller
2012

Canopy
Height
2011 (cm)

Canopy
Height
2012
(cm)
si
μ
g
.

Final
Biomass (g)

Cumulative
2012 (g)

Max.
Tiller
2011

Max.
Tiller
2012

Treatment

μ

sig.

μ

sig.

μ

sig.

μ

sig.

μ

sig.

μ

sig.

μ

sig.

Control

N/A

N/A

43.4

A

89

A

128

A

50

A

82

A

83.3

A

56.7

A

F1I40

31.6

A

42.4

A

90

A

147

A

58

A

94

A

50

B

44.5

B

F1I70

24.9

A

46.3

A

93

A

153

A

52

A

107

A

49.4

B

43.6

B

F3I40

31.1

A

46.5

A

103

A

142

A

59

A

81

A

49.4

B

44.9

B

F3I70

16

B

29.1

B

92

A

130

A

42

A

79

A

42.4

C

37.7

C

Pinegrass
Final
Biomass (g)

Cumulative
2012 (g)

Max.
Tiller
2011

Max.
Tiller
2012

Min.
Tiller
2011

Treatment

μ

sig.

μ

sig.

μ

sig.

μ

sig.

μ

sig.

μ

Control

N/A

N/A

5.2

AB

129

A

105

AB

33

A

F1I40

3.8

A

5.1

AB

140

A

123

A

35

A

F1I70

2.7

AB

5.4

A

142

A

127

A

31

F3I40

2.5

B

3.7

B

132

A

95

AB

F3I70

1.3

C

3.8

B

122

A

87

B

c.

Min.
Tiller
2012

Canopy
Height
2011 (cm)

Canopy
Height
2012 (cm)

sig.

μ

μ

14

AB

52.6

A

4.8

A

21

A

38.7

BC

39.5

BC

A

19

A

38.6

B

37.2

BC

35

A

15

A

32.3

CD

37

BC

26

A

7

B

31.5

D

33

C

sig.

sig.

Appendix 7: Summary of significance values and means for all vigour measurements in all three species. a.)
Bluebunch wheatgrass. b.) Rough fescue. c.) Pinegrass.

79