Targeted-Grazing As a Fuels Reduction Treatment: Evaluation of Vegetation Dynamics and Utilization Levels

Utah State University DigitalCommons@USU

All Graduate Theses and Dissertations Graduate Studies

5-2018

Targeted-Grazing as a Fuels Reduction Treatment: Evaluation of Vegetation Dynamics and Utilization Levels

Travis Decker Utah State University

Follow this and additional works at: https://digitalcommons.usu.edu/etd

Part of the Animal Sciences Commons

Recommended Citation Decker, Travis, "Targeted-Grazing as a Fuels Reduction Treatment: Evaluation of Vegetation Dynamics and Utilization Levels" (2018). All Graduate Theses and Dissertations. 7015. https://digitalcommons.usu.edu/etd/7015

This Thesis is brought to you for free and open access by the Graduate Studies at DigitalCommons@USU. It has been accepted for inclusion in All Graduate Theses and Dissertations by an authorized administrator of DigitalCommons@USU. For more information, please contact [email protected].

TARGETED-GRAZING AS A FUELS REDUCTION TREATMENT: EVALUATION OF

VEGETATION DYNAMICS AND UTILIZATION LEVELS

Travis Decker

A thesis submitted in partial fulfillment of the requirements for the degree

MASTER OF SCIENCE

Range Science

Approved:

______Eric Thacker, Ph.D. Beth Burritt, M.S. Major Professor Committee Member

______Corey Ransom, Ph.D. Mark R. McLellan, Ph.D. Committee Member Vice President for Research and Dean of the School of Graduate Studies

UTAH STATE UNIVERSITY Logan, Utah

2018

iii

ABSTRACT

Targeted-Grazing as a Fuels Reduction Treatment: Evaluation of Vegetation Dynamics and

Utilization Levels

Travis Decker, Master of Science

Utah State University, 2018

Major Professor: Dr. Eric Thacker Department: Wildland Resources

Wildfires have caused concern as they have increased in severity and intensity

over the last few decades. Land managers have sought management actions to mitigate

the risk of wildfire by reducing fuel loads, thus decreasing wildfire intensity. I conducted

two studies using grazing for fuel breaks. The first study was conducted at Camp

Williams, a National Guard camp near Bluffdale, Utah, where small arms and artillery training occurs. Managers at Camp Williams have created fuel breaks by implementing

targeted sheep and goat grazing to remove fine fuel and thin brush. Management

objectives set utilization of herbaceous fine fuels at 80% by weight. Questions arose as to

what the ecological impact prescribed grazing rates may have on native vegetation. I

evaluated three fuel breaks and quantified the impacts of targeted sheep and goat grazing

on vegetation dynamics. During the summer of 2015, herbaceous cover, shrub cover, and

shrub density was collected along eight paired (inside fuel break and outside fuel breaks) iv transects within each fuel break. The objectives of the second study were to determine how different levels of grazing utilization (30%, 50%, and 80%) relate to fuel characteristics and subsequent fire behavior. Results suggest that moderating grazing utilization levels (50%) may allow for more sustainable fuel reduction treatments while still reducing wildfire risks.

(73 pages) v

PUBLIC ABSTRACT

Targeted-Grazing as a Fuels Reduction Treatment: Evaluation of Vegetation Dynamics and

Utilization Levels

Travis Decker

Wildfires have caused concern as they have increased in severity and intensity

over the last few decades. Land managers have sought management actions to mitigate

the risk of wildfire by reducing fuel loads, thus decreasing wildfire intensity. Camp

Williams is a National Guard camp near Bluffdale, Utah, where small arms and artillery

training occurs. Managers at Camp Williams have created fuel breaks by implementing

targeted sheep and goat grazing to remove fine fuel and thin brush. Management

objectives set utilization of fine fuels (herbaceous) at 80% by weight. Questions arose

regarding the ecological impact of the prescribed grazing rates in these fuel breaks. This study evaluated three fuel breaks and quantified the impacts of targeted sheep and goat

grazing at 80% utilization. During the summer of 2015, herbaceous cover, shrub cover,

shrub density, and bunch grass density was collected along eight paired (inside fuel break

and outside fuel breaks) transects. Results indicate that the current management grazing

plan could lead to an increase of invasive annual grasses, which may be

counterproductive in fuel breaks. Often fine fuel treatments rely on high levels of grazing

utilization (> 80%). However, high levels of utilization can lead to ecological degradation

by reducing or eliminating native bunchgrasses. The objectives of the second study

conducted were to determine how different levels of grazing utilization (30%, 50%, and vi

80%) relate to fuel characteristics and subsequent fire behavior. Results suggest that moderating grazing utilization levels (50%) may allow for more sustainable fuel reduction treatments while still reducing wildfire risks.

vii

ACKNOWLEDGMENTS

I would like to thank my advisor Dr. Eric Thacker for the chance he gave me to learn and be a part of such a great project; I am truly grateful for it. I am grateful to Doug

Johnson and Sean Hammond of the Utah National Guard for helping me and guiding me throughout this project. I am thankful to Susan Durham for assisting me in my analysis.

Thanks to Utah State University along with USU Richmond Research Farm in

Richmond, UT. I am thankful for the USDA ARS Poisonous Plant Lab for letting me use their cattle for research. I personally would like to thank my family (my parents who keep pushing me) and my wife Shelby Decker for help and support. Along with amazing friends that have been there along the way. Funding was provided by the Utah National

Guard, Army Garrison Camp Williams.

Travis Decker

viii

CONTENTS

Page

ABSTRACT ...... iii

PUBLIC ABSTRACT ...... v

ACKNOWLEDGMENTS ...... vii

LIST OF TABLES ...... x

LIST OF FIGURES ...... xii

CHAPTERS

1. INTRODUCTION ...... 1

Problems and Objectives ...... 1 Human Influence on Wildfire Regimes ...... 2 Utah Wildfire History ...... 4 Camp W.G. Williams ...... 4 Wildland Urban Interface (WUI) ...... 6 Fuel Reduction ...... 7 Targeted Grazing ...... 8 Objectives ...... 10 References ...... 11 Tables and Figures ...... 16

2. EVALUATING TARGETED GRAZING AS A FUELS TREATMENT ...... 20

Abstract ...... 20 Problems and Objectives ...... 20 Methods ...... 22 Results ...... 25 Discussion ...... 26 Conclusion ...... 28 References ...... 28 Figures ...... 31

3. HOW ARE GRAZING UTILIZATION LEVELS AND FIRE BEHAVIOR RELATED? ...... 40

Abstract ...... 40 Problems and Objectives ...... 41 Methods ...... 42 Results ...... 43 Rate of Spread ...... 44 Discussion ...... 45 Conclusion ...... 47 References ...... 48 Tables and Figures ...... 50

4. CONCLUSION AND DISCUSSION ...... 53

Conclusion ...... 54

APPENDIX ...... 55

LIST OF TABLES

Table Page

1.1 Population in communities surrounding Camp Williams, UT, from 2000 to 2010 (U.S. Census)...... 16

1.2 Estimated costs for different fuel management options on rangeland settings. Table modified from Strand et al. 2014 ...... 17

3.1 Plot layout for grazing utilization trial at Richmond, UT ...... 50

3.2 Richmond fuel biomass averages collected post-grazing ...... 50

3.3 BeHavePlus5 inputs and data source for inputs. Dead Fuel Moisture of Extinction received from Utah State University/Admin...... 51

A.1 Cover percent and p-values of herbaceous plants along with bare ground and litter in Oak Springs. Showing between grazed (treatment) and ungrazed (reference) averages. (Other brush consists of Oregon grape, antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush, and common snowberry). Asterisks (*) shows difference at P=.05 level ...... 56

A.2 Cover percent and p-values of herbaceous plants along with bare ground and litter in Beef Hollow. Showing between grazed (treatment) and ungrazed (reference) averages. (Other veg. consists of jointed goatgrass, medusahead grass, pricklypear cactus. Other brush consists of Oregon grape, antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush). Asterisks (*) shows difference at P=.05 level ...... 56

A.3 Cover percent and p-values of herbaceous plants along with bare ground and litter in Wood Hollow. Showing between grazed (treatment) and ungrazed (reference) averages. (Other veg. consists of jointed goatgrass, pricklypear cactus. Other brush consists of antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush). Asterisks (*) shows difference at P=.05 level ...... 57

A.4 Density and p-values of brush/shrubs at Oak Springs. Other brush consists of (Oregon grape, antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush, and common snowberry). Asterisks (*) shows difference at P=.05 level ...... 57

A.5 Density and p-values of brush/shrubs at Beef Hollow. Other brush consists of (Oregon grape, antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush). Asterisks (*) shows difference at P=.05 level ...... 58

A.6 Density and p-values of brush/shrubs at Wood Hollow. Other brush consists of (Antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush). Asterisks (*) shows difference at P=.05 level ...... 58

A.7 Flame lengths at 8.05 kph in meters. Showing the control and the three different treatment means, standard error, and p-value. Asterisks (*) shows no difference between treatments at P=0.05 ...... 59

A.8 Flame lengths at 40.23 kph in meters. Showing the control and the three different treatment means, standard error, and p-value. Asterisks (*) shows no difference between treatments at P=0.05 ...... 59

A.9 Rate of spread in kph with 8.05kph mid-flame wind speeds. Showing the control and the three different treatment means, standard error, and p-value. Asterisks (*) shows no difference between treatments at P=0.05 ...... 60

A.10 Rate of spread in kph with 40.23kph mid-flame wind speeds. Showing the control and the three different treatment means, standard error, and p-value. Asterisks (*) shows no difference between treatments at P=0.05 ...... 60

xii

LIST OF FIGURES

Figure Page

1.1 Map produced by Scott Frost showing the 2010 Machine Gun Fire on Camp Williams that burned three houses in Herriman, Utah (Frost 2015) ...... 18

1.2 Map showing Camp Williams in relation to communities surrounding the camp (Camp W.G. Williams Joint Land Use Study Implementation) ...... 19

2.1 Camp Williams’ sheep and goat grazed fuel break pastures. Oak Spring, Wood Hollow, and Beef Hollow ...... 31

2.2 Percent brush cover inside (grazed) and outside (ungrazed) of fuel breaks at Oak Springs in 2015 on Camp Williams. Statistical differences (p = 0.05) indicated with different letters (Other* consist of Mahonia aquifolium, Purshia tridentate, Petradoria pumila, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa, Symphoricarpos albus) ...... 32

2.3 Percent ground cover inside (grazed) and outside (ungrazed) of fuel breaks at Oak Springs in 2015 on Camp Williams. Statistical differences (p = 0.05) indicated with different letters ...... 33

2.4 Density of brush inside (grazed) and outside (ungrazed) fuel breaks at Oak Springs in 2015 on Camp Williams. The statistical difference (p=0.05) indicated with different letters. (Other* consist of Mahonia aquifolium, Purshia tridentate, Petradoria pumila, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa, Symphoricarpos albus) ...... 34

2.5 Percent brush cover inside (grazed) and outside (ungrazed) of fuel breaks at Beef Hollow in 2015 on Camp Williams Statistical differences (p = 0.05) indicated shown with different letters. (Other* consist of Mahonia aquifolium, Purshia tridentate, Petradoria pumila, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa) ...... 35

2.6 Percent ground cover inside (grazed) and outside (ungrazed) of fuel breaks at Beef Hollow in 2015 on Camp Williams. Statistical differences (p = 0.05) indicated with different letters. (Other veg.* consist of Aegilops cylindrical, Taeniatherum caput-medusae, Opuntia polyacantha) ...... 36

2.7 Density of brush inside (grazed) and outside (ungrazed) of fuel breaks at Beef Hollow in 2015 on Camp Williams. The statistical difference (p=0.05) indicated with different letters. (Other* consist of Mahonia aquifolium, xiii

Purshia tridentate, Petradoria pumila, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa) ...... 37

2.8 Percent brush cover inside (grazed) and outside (ungrazed) of fuel breaks at Wood Hollow in 2015 on Camp Williams Statistical differences (p = 0.05) indicated with different letters. (Other* consist of Purshia tridentate, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa) ...... 38

2.9 Percent ground cover inside (grazed) and outside (ungrazed) of fuel breaks at Wood Hollow in 2015 on Camp Williams. Statistical differences (p = 0.05) indicated with different letters. (Other veg.* consist of Aegilops cylindrical, Opuntia polyacantha) ...... 39

2.10 Density of brush inside (grazed) and outside (ungrazed) fuel breaks at Wood Hollow in 2015 on Camp Williams. The statistical difference (p=0.05) indicated shown with different letters. (Other* consist of Purshia tridentate, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa) ...... 39

3.1 Modeled flame length (meters) with two different wind speed scenarios at 0%, 30%, 50% and 80% utilization rates. The dotted line represents a 1.22m flame length (4ft) where direct firefighting methods are no longer possible ...... 52

3.2 The Modeled rate of spread at kilometers per hour with two different wind speed scenarios at 0%, 30%, 50%, and 80% utilization rates ...... 52

CHAPTER 1

INTRODUCTION

Problems and Objectives

Wildfires are a natural process in rangelands throughout the western United

States. Wildfires are driven by three factors, topography, vegetation (fuel type and load),

and weather (temperature and humidity). Wildfires have become a higher risk due to an increase of urban development. Many urban areas have been developed within sagebrush

and oakbrush communities in Utah. Oak brush (Quercus gambelii) communities are reported to have a wildfire return interval between 35 and 100 years.1 Historically,

wildfire return intervals in Wyoming big sagebrush (Artemisia tridentata wyomingensis)

communities are between 50 to 240 years.2,3 Wildfire history on Mountain big sagebrush

(Artemisia tridentata vaseyana) has a fire frequency from 3 years to as long as 200

years.2,3,4 However, it is difficult to determine exact historic wildfire regimes in

sagebrush systems as wildfires burn most vegetation. Retrieving historic wildfire scars to

obtain wildfire frequency is difficult due to the lack of trees found in sagebrush

communities.

Large wildfires that occurred in the late 1800s and early 1900s, such as the

Santiago Canyon Fire in California, caused land managers and government officials to

increase wildfire suppression in the western U.S.5 In 1934, the policy was to extinguish

any wildfire by 10:00 a.m. the following day. Fire suppression lead to a decrease of wildfires and an increase of fuels. By the 1960s and 1970s, multiple policies started to change, as evidence increased suggesting wildfires do need to occur in different 2

ecosystems. Since 1983, wildfires in the western U.S. have increased in total hectares

burned and number of wildfires.5 The Yellowstone fires in 1988 showed that years of

wildfire suppression increased the amount of fuels that had built up causing wildfires to

burn out of control. Yet, even with new policies, Stephens and Ruth reported that

approximately 97-99% of wildfires are suppressed on the first initial attack.6

Human Influence on Wildfire Regimes

In the late 1800s and into the early 1900s over grazing became a problem in the western United States. Open ranges allowed livestock to graze across rangelands throughout the west. It is estimated that there were 26 million cattle and 20 million sheep in 17 western states by the 1890s.7 In Utah population of sheep was 3 million before

declining after the mid-1930s.8 The Taylor Grazing Act was passed in 1934 to help

regulate grazing. These regulations assisted in improving the rangelands into a more

sustainable resource. Shortly after the Taylor Grazing Act was passed, livestock numbers

decreased. Sheep population numbers declined quickly as majority of sheep producers

switched to a cattle operation.9

A study conducted in Nevada in the early 1900s measured a site dominated by

bunchgrass and desirable shrubs before grazing by livestock. Monitoring occurred at the same site 50 years later after heavy grazing by livestock. There was a loss in desirable shrubs, an increase in erosion, and less than 5% bunchgrass cover. The site was dominated by cheatgrass (Bromus tectorum).10 Cheatgrass is a nonnative annual grass

introduced into the U.S. in the late 1800s. Cheatgrass increased due to overgrazing and 3

caused a decrease in native vegetation which lead to disturbed rangelands in the early

1900s.11,12

Cheatgrass germinates in late fall, goes dormant over winter, and increases

growth in the spring as temperatures rise. It is hard to predict the annual cheatgrass

biomass production because it varies from year to year. It does well on disturbed sites or

on shallow soils where native vegetation lacks the ability to successfully compete for

moisture and nutrients.11,13 Cheatgrass tends to dry out quicker than native bunch grasses,

maturing by June whereas bunch grasses typically stay green throughout July.14

The early maturing of cheatgrass provides a dry, fine continuous fuel that ignites

easily and can spread wildfires rapidly. The dead biomass from cheatgrass can cause

wildfires to occur earlier in the fire season than typical native vegetation. Because fire

occurs earlier in the season on cheatgrass invaded sites, it can damage native vegetation,

making it difficult for native vegetation to recover.11 Cheatgrass has increased the

frequency and size of wildfires.10 A study conducted in the Idaho plains found within 3 to

5 years post wildfire, the same area is susceptible to another wildfire due to new

cheatgrass and litter from cheatgrass.15 By the 1930s, cheatgrass became established in its

current range in the western United States.11 It has been reported that 39 of the 50 largest western wildfires in 2000 to 2009 were ignited in cheatgrass dominated sites. In that same time period, cheatgrass burned twice as much as any other vegetation.16

The increase of cheatgrass and the resulting increase of wildfires has led to a

decrease in native vegetation. Frequent wildfires destroy sagebrush and sagebrush

seedlings, which leaves disturbed areas for cheatgrass to invade. Wyoming big sagebrush 4 can take 50 to 120 years for recovery post wildfire, while mountain big sagebrush may take 35 to 100 years to recover. When sagebrush dominated rangelands become infested with cheatgrass, these sites become susceptible to more frequent wildfires.3

Utah Wildfire History

The major range types in Utah include salt desert flats, sage-steppe, and upland forest. This research focuses on Wyoming Big and Mountain Big Sagebrush communities along with Gambel Oak rangelands. The wildfire return interval for these communities is estimated to be 10 to 300 years.1,17 Wildfires in Utah have increased in recent years. The

Milford Flats Fire started on July 6, 2007, in Beaver County, near Milford, Utah. This wildfire is the largest recorded in the state of Utah, burning a total of 147,207 ha and causing an auto accident that claimed two lives.18,19 From 1996 to 2012, a total of 15 large wildfires occurred that burned a total of 534,594 ha.19

Camp W.G. Williams

Camp W.G. Williams is a Utah National Guard training ground in central Utah located between Bluffdale and Saratoga Springs. Camp Williams was officially established in 1928, though before this time, it was used for training by local military units. Camp Williams consists of 9,712.5 ha used for training exercises in combat scenarios. Trainings consist of small arms weapons, demolition and grenade ranges, and artillery ranges. All live fire ranges occur in the western end of the camp in what is 5

designated as the impact zone. Multiple military units from around the country and parts

of the world come to Camp Williams to train, making training a vital importance.20

Camp Williams has a 30-year average of 355.6 mm of precipitation per year.

Native vegetation on Camp Williams is dominant sagebrush at the lower elevations (1300

m) and a mix of sagebrush and oak brush at the higher elevations (2300 m). Grasses

consist of native bunch grasses such as Sandberg bluegrass (Poa secunda), needle and thread (Hesperostipa comata), western wheatgrass (Pascopyrum smithii), and blue bunch wheatgrass (Pseudoroegneria spicata). Camp Williams has many invasive grasses as well, such as cheatgrass, medusahead (Taeniatherum caput-medusae), jointed goatgrass

(Aegilops cylindrical), and bulbous bluegrass (Poa bulbosa) along with many non-native weedy forbs.

Research from Frost (2015), showed from 1985 to 2012, Camp Williams had a total of 86 wildfires that burned 12,279 ha.21 Out of the 86 wildfires, 18 of those wildfires

burned more than 40 ha since 1985.21 Wildfires on the camp typically burn between mid-

June to October. Frost found that wildfires around 40 ha occur every one-two years,

while fires ≥400 ha occur about once every 4 years.21 In the past 15 years, at least four

different training occasions have caused wildfires that threatened homes.22 The “Machine

Gun Fire” was one of the more recent fires and started during training exercises with a

0.50 caliber machine gun. A bullet round sparked a fire that started in the southwestern

portion of the camp. It spread north to northeast through the camp and burned across

multiple fuel breaks, fire lines, and roads before leaving the camp (Fig. 1.1). This fire 6 caused an estimated 5,000 people and 1,652 homes to be evacuated before destroying three homes in Herriman, Utah, and burning a total of 1,761 ha of rangeland.21,22

In recent years, urban development around Camp Williams has increased. There are four main communities that border Camp Williams: Eagle Mountain, Saratoga

Springs, Herriman, and Bluffdale (Fig. 1.2).23 Total population growth around the camp from 2000 to 2010 increased by 60,000 people (See Table 1.1).24

In response to urban growth around Camp Williams, managers have implemented fuel treatments to create a defendable space between the camp boundaries and the urban development surrounding the camp.

Wildland Urban Interface (WUI)

Wildland Urban Interfaces (WUI) are populated communities bordering undeveloped natural environments that are prone to wildfires. Since the 1960s, populations have increased in WUI communities from 25 million to 140 million people in the United States.25 When wildfires interact with urban development, the situation can become more complex. Fighting wildfires within WUIs increases the cost of wildland fire suppression.26 On an average year, there are more than 2,600 structures lost due to wildfires nationwide in WUI areas.27

Suppression cost of wildfires increases due to multiple factors, one of which is the ability to contain and/or control wildfires before they enter into urban areas. Defending an urban area requires more firefighters and thus increase demand for firefighting resources. Gorte (2013) reported that since 2002 an average of $3 billion per year is spent on wildfire suppression. This has increased from the $1 billion average in the 1990s.26 7

Rasker (2008) conducted a study in Montana that concluded if wildfires threaten to burn

150 homes, the cost of suppression increases to $13 million per year. This averages the

cost of suppression of WUI wildfires around $1 million per fire.28 Research conducted in

Oregon and California show that wildfire suppression costs increase with the growing

amount of structures and their relative location (distance) to the wildfire.29 The State of

Utah paid around $50 million in suppression costs for wildfires in 2013.30

Fuel Reduction

Rangeland managers control fuels to reduce the risk of wildfires. Different

treatment types to manage fuel loads include: chemical control, mechanical, prescribed

burns, and cultural.

Herbicides may kill specific plants, yet this does not remove the plant matter,

leaving fuel for wildfires to burn. Herbicide treatments may cost $61 to $617 per

hectare.31 Mechanical control can remove the vegetation. Mechanical treatment from

human operation are the costliest with treatments rates of $864 to $1,976 per hectare.

Additionally, some mechanical treatments such as mowing can cause sparks that may ignite wildfires.

Herbicide treatments are considered a high cost treatment on Camp Williams.

Mechanical treatments do occur on a small scale on the camp, but as stated in

Mendenhall 2004, this is costly and publicly undesirable.32 Prescribed burning on Camp

Williams is not allowed according to a report by Mendenhall (2004).32 At about this same time in 2004 Camp Williams implemented the use of grazing to create fuel breaks.

Targeted grazing treatments tend to be more acceptable to the general public.32 Targeted 8

grazing may be more acceptable yet is still expensive due to the cost of equipment,

setting up and taking down electric fence, hauling livestock, animal care, and hauling water for the livestock (Table 1.2).

Targeted Grazing

Targeted grazing is a form of grazing that targets a specific area or a species of vegetation. The type of livestock, intensity of grazing, and timing of grazing are important factors to consider when implementing targeted grazing.33 Cattle are more

effective at consuming grasses, whereas sheep can be more effective consuming leafy

vegetation such as multiple types of forbs. Sheep still forage on grasses as well.34 Nader

et al. (2007) reviews different factors managers should consider when using targeted

grazing as a fuel management tool. These factors include species of livestock grazed

(cattle, sheep, goats, or a combination); the animals’ previous grazing experience along

with time of year as it relates to plant physiology. Stock density and duration also are

factors in the success of fuels reduction in a treated area.31

Targeted grazing can achieve many different objectives, one such outcome of

targeted grazing is fuels reduction. Grazing occurs in defined places on the landscape,

such as bordering an urban development. As grazing reduces the level of fuels, wildfires

are less intense and have slower progression. Less intense, slower spreading wildfires

tend not to burn all vegetation, leaving the possibility for native vegetation to regrow post

fire. The slowing of a wildfire can lead to a defensible space or give time for firefighters

to attack and stop the progression of the wildfire. 9

Targeted grazing can be used to create fuel breaks. A fuel break is a section or

strip of land where fuel is reduced or removed to slow or stop the progression of

wildfires.35 By reducing fuel loads and altering fuel continuity, targeted grazing can reduce wildfire intensity and rate of spread on rangelands.31,36,37,38,39,40,41,42,43 When fuel

breaks are in a cheatgrass dominated community, removing 80% of the fuel load has been

shown to decrease wildfire spread and flame length.36,40 These studies suggest that heavy

grazing (80% utilization) on cheatgrass dominated sites may reduce cheatgrass

abundance.37

Research on grazed fuel breaks tends to recommend high utilization rates around

80%. However, Davies et al. (2015) showed that grazing at lower utilization levels can be

effective in controlling wildfires.41,42 Furthermore, moderate rates of grazing (40-50% utilization) provides stability for native vegetation.44 Little research has been done

examining the relationship between utilization rates and fire behavior. If a moderate

utilization rate can create similar wildfire behavior as an 80% utilization treatment, land

and fuel managers can maintain native vegetation by using the lower utilization rate in

treatment areas.

Studies have shown that goats can be used to treat and reduce shrubs and brush

type fuels (1 hr and 10 hr fuels).31,39 In another study conducted in Oregon, researchers

found that dormant winter grazing at 60% biomass removal reduced wildfire risk in

sagebrush communities.41,42 This study later applied a control burn to the site and found

that grazed areas had less fuel and increased fine fuel moisture, resulting in lower flame

lengths and rate of spread. Grazed/burned areas also showed less damage to native 10

vegetation, leaving some native bunchgrasses unburned, compared to the ungrazed site

where the control burn had higher intensities that removed all vegetation.42

However, managers should recognize the limits of grazing treatments. Nader et al.

(2007) suggested that grazing can only effect fine fuels in the 1-hr and 10-hr fuel range,

or vegetation smaller than 2.54 cm in diameter.31 While only fine fuels are treated, this treatment is still effective to reduce the risk of wildfires. Timing of grazing is important; therefore, if management objectives are to remove annual grasses it may be better to graze in the late fall and when annual grasses are germinating, or graze annual grasses in the early spring prior to perennial grasses reaching the boot stage.45 There are concerns about spring grazing since native bunchgrasses are sensitive to spring grazing especially once the perennial grasses have booted. Concerns about targeted grazing as a fuel treatment are that overgrazed areas damage the landscape. Excessive grazing can lead to desertification and soil compaction. Fall grazing, especially in cheatgrass dominate sites,

is recommended. Schmelzer et al. (2014) state that they were able to reduce cheatgrass

fuels while causing no risk to the livestock or plant community.43

Camp Williams uses livestock grazing to manage fine fuels and brush to create

fuel breaks. Camp Williams implemented the use of goats and sheep in 2004 to create

fuel breaks in oakbrush communities. In 2013, Camp Williams implemented the use of

cattle grazing to reduce fine fuel loads.

Objectives

Excessive grazing can increase weedy species such as cheatgrass.44 With the

increase of cheatgrass, the threat of wildfire ignition and spread increases.10 Camp 11

Williams implemented its targeted grazing efforts in 2004; managers are interested on the impacts of targeted grazing to create fuel breaks. The questions I will answer are: 1- Has cheatgrass increased or decreased due to current grazing practices? 2- Has native vegetation persisted through this grazing treatment? 3- Can lower utilization rates still reduce wildfire risk? The following chapters follow “Rangelands” guidelines.

References

1. SIMONIN, K. A. 2000. Quercus gambelii. Fire Effects Information System,[Online]. US Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Available: http://www.fs.fed.us/database/feis/ [2015, October 28].

2. MILLER, R.F., AND R.J. TAUSCH. 2000. The role of fire in pinyon and juniper woodlands: a descriptive analysis. In Proceedings of the invasive species workshop: the role of fire in the control and spread of invasive species. Fire conference 15-30.

3. BAKER, W.L. 2006. Fire and restoration of sagebrush ecosystems. Wildlife Society Bulletin 34.1.

4. MILLER, R., C. BAISAN, J. ROSE, AND D. PACIORETTY. 2001. Pre-and post-settlement fire regimes in mountain big sagebrush steppe and aspen: the northwestern Great Basin. Final report, 28.

5. U.S. Forest Service Fire Suppression. U.S. Forest Service History. Forest History Society,www.foresthistory.org/ASPNET/Policy/Fire/Suppression/Suppression.asp x Web. 20 Feb. 2016.

6. STEPHENS, S.L., AND L.W. RUTH. 2005. Federal forest-fire policy in the United States. Ecological Applications 15(2):532-542.

7. MCGINTY, E.L., B. BALDWIN, AND R. BANNER. 2009. A review of livestock grazing and range management in Utah. Setting the Stage for a Livestock Grazing Policy in Utah.

8. FORREST, T. History of Grazing in Utah. Utah Department of Agriculture and Food. State of Utah. http://ag.utah.gov/conservation-environmental/grazing- improvement-program/history-of-grazing-in-utah/75-conservation-and- environmental/grazing.html Web. 17 Jan. 2016.

9. GODFREY, E.B., 2008. Livestock grazing in Utah: history and status. A report for the Utah Governor’s Public Lands Policy Coordination Office. Department of Applied Economics, Utah State University.

10. BILLINGS, W.D. 1994. Ecological impacts of cheatgrass and resultant fire on ecosystems in the western Great Basin. SB Monsen and SG Kitchen, 22-30.

11. PELLANT, M., 1996. Cheatgrass: the invader that won the west. US Department of the Interior, Bureau of Land Management. 22.

12. KNAPP, P.A., 1996. Cheatgrass (Bromus tectorum L) dominance in the Great Basin Desert: history, persistence, and influences to human activities. Global Environmental Change 6(1):37-52.

13. ZOUHAR, K. 2003. "Bromus tectorum." Fire Effects Information System. Fort Collins, CO: USDA Forest Service, Rocky Mountain Research Station Fire Sciences Laboratory. http://www. fs. fed. us/database/feis/. Author: Michele A. James Scanned Images: Dave Egan Reviewers: Dave Brewer, Dave Egan, Chris McClone, and Judy Springer Series Editor: Dave Egan. [2016, February 23].

14. YOUNG, J.A. AND R.R. BLANK. 1995. Cheatgrass and wildfires in the intermountain west. California exotic pest plant council symposium proceedings. 1-3.

15. WHISENANT, S.G., 1990. Changing fire frequencies on Idaho's Snake River Plains: ecological and management implications. Changing fire frequencies on Idaho's Snake River Plains: ecological and management implications., (INT-276), 4-10.

16. BALCH, J.K., B.A. BRADLEY, C.M. D'ANTONIO, AND J. GOMEZ‐DANS. 2013. Introduced annual grass increases regional fire activity across the arid western USA (1980–2009). Global Change Biology 19(1):173-183.

17. MCGINTY, E.I. AND C.M. MCGINTY. 2009. Rangeland resources of Utah/fire in Utah. Utah State University Cooperative Extension Service. Logan, UT.

18. TANNER, T. "Scarred Land Recovering 5 Years after Milford Flat Wildfire." Fox 13 Salt Lake City. Fox 13 Salt Lake City, 24 July 2012. http://fox13now.com/2012/07/24/millford-flat-wildfire-five-years-of-restoration/ Web. 4 Feb. 2016.

19. University of Utah, Utah State University, Weber State University. An Analysis of a Transfer of Federal Lands to the State of Utah / Ch. 9 Wildfire In Utah. Bureau of Economic and Business Research, 2014.

20. Camp W.G. Williams. Camp W.G. Williams. Utah National Guard. https://www.ut.ngb.army.mil/campwilliams/ Web. 12 Nov. 2015.

21. FROST, S.M. 2015. Fire Environment Analysis at Army Garrison Camp Williams in Relation to Fire Behavior Potential for Gauging Fuel Modification Needs. Logan, UT, Utah State University.

22. The Salt Lake Tribune "Machine Gun Fire Not the First at Camp Williams." The Salt Lake Tribune. 26 Sept. 2010. http://archive.sltrib.com/article.php?id=50319625&itype=CMSID Web. 12 Feb. 2016.

23. "Camp W.G. Williams Joint Land Use Study Implementation." Camp W.G. Williams Joint Land Use Study Implementation. Matrix Design Group http://www.campwilliamsjlus.com/index.php/project-overview/project-study-area. Web. 4 May 2016.

24. U.S. Census, “Census of Population and Housing”. https://www.census.gov/ . Retrieved June 4, 2015

25. BAILEY, D.W., 2007. The wildland/urban interface crisis, is there a solution?, 4th International Wildland Fire Conference.

26. GORTE, R. AND Headwater Economics. 2013. The rising cost of wildfire protection. Headwaters Economics.

27. National Interagency Fire Center. NIFC. https://www.nifc.gov/ Web. 14 Nov. 2015.

28. RASKER, R., and Headwaters Economics. 2008. Montana Wildfire Cost Study Technical Report.

29. GUDE, P.H., K. JONES, R. RASKER, AND M.C. GREENWOOD. 2013. Evidence for the effect of homes on wildfire suppression costs. International Journal of Wildland Fire 22(4):537-548.

30. JUDD, B. "2012 Wildfire Season a Destructive One in Utah; Will 2013 Be as Bad?" Deseret News. Deseret News, http://www.deseretnews.com/article/865581710/2012-wildfire-season-a- destructive-one-in-Utah-will-2013-be-as-bad.html 14 June 2013. Web. 5 Feb. 2016.

31. NADER, G., Z. HENKIN, E. SMITH, R. INGRAM AND N. NARVAEZ. 2007. Planned Herbivory in the Management of Wildfire Fuels: Grazing is most effective at treating smaller diameter live fuels that can greatly impact the rate of spread of a fire along with the flame height. Rangelands 29(5):18-24.

32. MENDENHALL, M. 2004. "Fighting Fire with Goats on Camp Williams." Utah Forest News. Utah State University Extension. https://forestry.usu.edu/files- ou/UFN0804.pdf#page=4 Web. 25 Jan. 2016.

33. LAUNCHBAUGH, K. AND J.W. WALKER. 2006. Targeted grazing—a new paradigm for livestock management. Targeted grazing: a natural approach to vegetation management and landscape enhancement. Centennial, CO, USA: American Sheep Industry Association, 2-8.

34. BURRITT, E. AND R. FROST. 2006. Animal behavior: principles and practices. Targeted grazing: a natural approach to vegetation management and landscape enhancement. American Sheep Industry Association, Cottrell Printing, Centennial, Colo, 22-31.

35. NRCS Fuel Breaks. National Resource Conservation Service (2005, May). Retrieved from http://www.nrcs.usda.gov/Internet/FSE_DOCUMENTS/nrcs143_026161.pdf. Web. 20 Jan. 2016.

36. DIAMOND, J.M., 2009. Effects of targeted grazing and prescribed burning on fire behavior and community dynamics of a cheatgrass (Bromus tectorum)- dominated landscape. Utah State University.

37. STRAND, E.K., K.L. LAUNCHBAUGH, R.F. LIMB, AND L.A. TORELL. 2014. Livestock grazing effects on fuel loads for wildland fire in sagebrush dominated ecosystems. Journal of Rangeland Applications 1:35-57. 15

38. DAVIES, K.W., J.D. BATES, T.J. SVEJCAR, AND C.S. BOYD. 2010. Effects of long-term livestock grazing on fuel characteristics in rangelands: an example from the sagebrush steppe. Rangeland Ecology & Management 63(6):662-669.

39. TAYLOR JR, C.A., 2006. Targeted grazing to manage fire risk. Targeted grazing: A natural approach to vegetation management and landscape enhancement. 107- 112.

40. SCHMELZER, L., 2009. Reducing fuel load of key cheatgrass (Bromus tectorum) dominated range sites by the use of fall cattle grazing. Reno, NV: University of Nevada.

41. DAVIES, K.W., C.S. BOYD, J.D. BATES, AND A. HULET. 2015. Dormant season grazing may decrease wildfire probability by increasing fuel moisture and reducing fuel amount and continuity. International Journal of Wildland Fir 24(6):849-856.

42. DAVIES, K.W., C.S. BOYD, J.D. BATES, AND A. HULET. 2016. Winter grazing can reduce wildfire size, intensity and behaviour in a shrub- grassland. International Journal of Wildland Fire 25(2):191-199.

43. SCHMELZER, L., B. PERRYMAN, B. BRUCE, B. SCHULTZ, K. MCADOO, G. MCCUIN, S. SWANSON, J. WILKER, AND K. CONLEY. 2014. Case Study: Reducing cheatgrass (Bromus tectorum L.) fuel loads using fall cattle grazing. The Professional Animal Scientist 30(2):270-278.

44. LOVINA, R., K. LAUNCHBAUGH, J. JONES, L. BABCOCK, K.R. AMBROSE, A. STEBLETON, T. BREWER, K. SANDERS, J. MINK, AND G. HYDE. 2009. Rangelands: An introduction to Idaho wild open spaces. Department of Rangeland Ecology and Management and Idaho Rangeland Resource Commission.

45. MOSLEY, J.C. AND L. ROSELLE. 2006. Targeted livestock grazing to suppress invasive annual grasses. Targeted Grazing: a Natural Approach to Vegetation Management and Landscape Enhancement’. (Ed. K Launchbaugh) 68-77.

Tables and Figures

Table 1.1. Population in communities surrounding Camp Williams, UT, from 2000 to 2010 (U.S. Census).

Human Population Growth Cities Population in 2000 Population in 2010 Eagle Mountain 2,157 21,415 Saratoga Springs 1,003 17,781 Herriman 1,523 21,785 Bluffdale 4,700 7,598 Total 9,383 68,579

Table 1.2. Estimated costs for different fuel management options on rangeland settings. Table modified from Strand et al. 2014.

Cost Treatments Source Description $/hectare $168- Herbicide Wolcott et al. 2007 Grass/Shrub/Tree Intermixed $2,471+ Torell et al. 2005 Sagebrush, treat with tebuthiron Aerial $51 Taylor et al. 2013 Sagebrush, treat with tebuthiron Aerial $76 Taylor et al. 2013 Tebuthiuron, ground application $128 Nader et al. 2007 Grass/Shrub $61-$671 Hand Crew Dan Macon Brush Removal $864-$1,482 Mechanical Nader et al. 2007 Mowing grasslands $61-$98 Wolcott et al. 2007 Mowing grasslands/shrubs $86-$1,235 Prescribed Fire Nader et al. 2007 Brush, range, and grasslands <$370 Cleaves et al. 2000 Brush, range, and grasslands $140 Mercer et al. 2007 Southeast U.S. $27-$850 Taylor et al. 2013 Healthy Sagebrush, perennial Understory $49 Taylor et al. 2013 Pinyon-Juniper with mature Shurbs $113 Combined Taylor et al. 2013 Closed-canopy pinyon-juniper Treatments brush management, herbicide, $506 and reseeding annual grass Taylor et al. 2013 Dominated; prescribed fire, $407 herbicide, and reseeding Targeted Nader et al. 2007 Targeted grazing with goats $148-$172 Grazing Dan Macon California sheep and goat grazing contractor <8.09 hectares $741 >8.09 hectares $370-$494 Varelas 2012 Targeted grazing with cattle $111-$160

Figure 1.1. Map produced by Scott Frost showing the 2010 Machine Gun Fire on Camp Williams that burned three houses in Herriman, Utah (Frost 2015).

Figure 1.2. Map showing Camp Williams in relation to communities surrounding the camp (Camp W.G. Williams Joint Land Use Study Implementation).

CHAPTER 2

EVALUATING TARGETED GRAZING AS A FUELS TREATMENT

Abstract

Wildfires have caused concern as they have increased in severity and intensity over the last few decades. Land managers have sought management actions to mitigate the risk of wildfire by reducing fuel loads, thus decreasing wildfire intensity. Camp

Williams is a National Guard camp near Bluffdale, Utah. Managers at Camp Williams have created fuel breaks as part of their wildfire mitigation plan using targeted sheep and goat grazing to remove fine fuel and thin brush. Management objectives set grazing of fine fuels at 80% utilization. Questions arose as to the ecological impacts of prescribed grazing rates within these fuel breaks. This study evaluated three fuel breaks and quantified the impacts of targeted sheep and goat grazing at 80% utilization. Results showed how invasive annual grasses, such as cheatgrass (Bromus tectorum) and bulbous bluegrass (Poa bolbosa) responded to heavy utilization rates. During the summer of 2015 data was collected on herbaceous cover, shrub cover and shrub density, along eight paired (inside fuel break and outside fuel breaks) transects.

Problems and Objectives

The western United States has seen an increase in wildfire size and frequency in recent years. This can be linked to the changes in fuel structure in rangelands throughout the west. Two main factors have led to the change in fuel buildup. First, many years of fire suppression have led to the accumulation of fuels1; and second, the increase of 21

cheatgrass (Bromus tectorum).2,3 As cheatgrass abundance increases, it changes fuel loads

and increases continuity of the fuel load, by filling in interspaces between grasses and

shrubs. This continuity can also increase the rate of wildfire spread.3

Fuels managers have struggled in recent decades over methods to control fuels. In

the last 15 years, grazing has become a popular and feasible option to reduce fuels organically and it has generally been embraced by the public living near the WUI.

Several studies have shown grazing can reduce wildfire risk. For example, in Carson

City, Nevada, sheep were used to remove cheatgrass to create fuel breaks; goats have been used in chaparral areas in Arizona and California.4 These studies often recommend

utilization rates around 80% to create fuel breaks.5,4,6 Diamond et al. (2009) showed that

80% utilization is effective in reducing the spread of wildfires and flame length in

cheatgrass communities. Diamond et al. (2009) also reported a decrease in cheatgrass

cover.

However, Young et al. (1987) reported that cheatgrass increases to fill the void

due to loss of perennial vegetation from excessive grazing.2,3 An increase of cheatgrass

can negatively affect rangelands and increase wildfire risk. Researchers have suggested

that overgrazing occurs when grazing exceeds the capacity of plants to recover.7 Most

rangelands can withstand 40-50% removal of native vegetation without damaging the

individual plants. If plants are grazed above 50% utilization at the wrong time of year,

grazing can damage plants which will lead to a decline of rangeland health. Cheatgrass is

a problem on rangelands because it has changed the fire return interval and increased the

spread of wildfires.3 This has caused wildfires to occur two months earlier and extends 22

the wildfire season.8 An earlier wildfire season can damage native vegetation and limit native vegetation from recovering. This will only further exacerbate cheatgrass

dominance. Whisenant (1990) reported that within 3 to 5 years post wildfire, the same

area in Idaho is prone to burn again due to cheatgrass introduced changes to fuel

characteristics.9 Cheatgrass promotes wildfire spread and frequency through the increase

of fine continuous fuel buildup.

Land managers and conservationists have expressed concern over the heavy

utilization levels typically recommended by previous research (Diamond et al. 2011) and

the possibility of converting perennial grass stands into cheatgrass dominated

communities. Therefore, the objective of my study is to evaluate fuel breaks at Camp

Williams to determine if targeted grazing to create fuel breaks has increased cheatgrass or other annual grasses or aided in the decline in native bunch grasses. This research will provide a long-term evaluation of the impacts of targeted grazing as a tool to create fuel breaks.

Methods

Camp Williams is a Utah National Guard training grounds and military installation in central Utah. Wildfires pose a threat to the camp’s facilities and more importantly threaten civilian lives and property surrounding Camp Williams. As of 2010, the total population of the four cities surrounding Camp Williams is 68,579.10

Average wildfires approximately 400 ha in size occur on Camp Williams around every four years11 and threaten the surrounding communities. Rangeland managers have

implemented various types of fuel breaks along camp boundaries to protect surrounding 23

communities. Since 2004, livestock producers have been contracted by Camp Williams to

strategically graze livestock in order to create fuel breaks.

Camp Williams consists of 9,712.5 ha of rangeland in central Utah. Camp

Williams’ average rainfall is 356 mm of rain in 30 years (1985-2015).12 Camp Williams’

vegetation consist of sagebrush (Artemisia tridentata) in lower elevations (1300 m) and a mix of sagebrush and oak brush (Quercus gambelii) at higher elevations (2300 m). Native bunchgrasses found on the camp includes Sandberg bluegrass (Poa secunda), needle and thread (Hesperostipa comata), western wheatgrass (Pascopyrum smithii), and bluebunch wheatgrass (Pseudoroegneria spicata). Invasive plants at Camp Williams includes cheatgrass (Bromus tectorum), medusahead (Taeniatherum caput-medusae), jointed goatgrass (Aegilops cylindrica), and bulbous bluegrass (Poa bulbosa), along with many

introduced weedy forbs such as Canada thistle (Cirsium arvense), musk thistle (Carduus

nutans) and common storksbill (Erodium cicutarium).

Since 2004, an average of 1,200 sheep/goat are used each year to create fuel

breaks on Camp Williams. Vegetation sampling was conducted in fuel breaks at Oak

Springs, Wood Hollow, and Beef Hollow in July 2015 (Fig. 2.1).

Oak Springs is dominated by a Wyoming big sagebrush (Artemisia tridentata var.

wyomingensis) community with some oak brush (Quercus gambelii). Oak Springs was grazed annually beginning in 2006 during the end of May. The Oak Springs fuel break is

12.6 ha (2.6 km X 60 m wide). Average stocking rate at Oak Springs was 10.79 AUM/ha.

The Beef Hollow fuel break is dominated by oak brush and covers 27.5 ha (5.5 km X 60

m). Sheep grazed Beef Hollow annually beginning in 2011 at the end of June. Average 24

stocking rate at Beef Hollow was 6.11 AUM/ha. The Wood Hollow fuel break is 40 ha (8

km X 60 m) and dominated by oak brush. Grazing has occurred annually in Wood

Hollow since 2004 typically during mid-July for about 2 weeks. Average stocking rate at

Wood Hollow was 4.98 AUM/ha. This study only used 5 km of the fuel break due to

cattle grazing occurring in a 3 km portion of the fuel break in 2015. Grazing occurred at

each site until the pasture reached 80% utilization.

Each fuel break was monitored using eight treatment/reference pairs (16 total).

Thirty meter transects were randomly placed inside of the fuel breaks (oriented the same direction as the fuel break) with a 10m buffer from the edge. Reference transects were

placed randomly within 100 m of the fuel breaks and ran parallel to the fuel breaks.

Canopy cover was collected for cheatgrass, bulbous bluegrass, native bunchgrasses,

native forbs, and introduced forbs. (All other invasive grasses detected were placed into

“other grasses” due to their low abundance). Litter and bare ground were also estimated

using the line-point intercept method. Shrub canopy cover was measured using the line

intercept method. Shrub density (sagebrush, oak brush, and “other brush”) was

determined by counting plants in 30 m x 1 m belt transects.17 Monitoring occurred in

May/June of 2015.

Data were analyzed using R Studio to perform a paired t-test on all transects sites

showing a difference at a P=0.05 level.15

During the analysis, I did not make pasture comparisons; I only compared grazed/ungrazed within each pasture due to vegetation community differences between pastures as well as the timing of grazing and duration of grazing and season of grazing. 25

Results

All sites exhibited little change when compared to ungrazed reference sites. In

Oak Springs, sagebrush cover and density in grazed areas were lower (P=0.0001) (Fig.

2.2 and 2.4). Native forbs and introduced forbs were higher in grazed sites (P=0.019 and

P=0.0001, respectively) (Fig. 2.3). Bulbous bluegrass was the only grass at the Oak

Springs that increased in the fuel break (grazing), however, it was still less than 5% cover

(P=0.049) (Fig. 2.3). The bare ground cover was higher while litter cover was lower

(P=0.0001 and P=0.018 respectively) in grazed pastures. In Oak Springs, density and

cover of oakbrush and all other brush were similar. There were no differences in

cheatgrass cover between the reference and grazed sites (Fig. 2.3).

There was more bulbous bluegrass in Beef Hollow in grazed pastures when

compared to reference pastures (P=0.0136) (Fig.2.6). Native bunchgrasses showed 50%

less abundance in the grazed site compared to the reference (P=0.0164) (Fig. 2.6). Litter

was less in grazed sites compared to reference sites (P=0.0376). Beef Hollow was the

only site to have trace amounts of medusahead (Taeniatherum caput-medusae) and

jointed goatgrass (Aegilops cylindrical), but there were not differences in either species

between grazed and ungrazed sites (Fig. 2.6). Beef Hollow showed no change in shrub

cover or density (Fig. 2.5 and 2.7)

In Wood Hollow shrub cover, shrub density, and herbaceous cover was very

similar when comparing grazed and reference sites (Fig 2.8, 2.9. and 2.10) However,

Wood Hollow had abundant cheatgrass cover in both control and grazed treatments (Fig. 26

2.9). Bulbous bluegrass cover was below 5% and was similar in grazed and ungrazed sites (See Appendix for all tables).

Discussion

In this study, none of the sites showed a significant increase in cheatgrass cover.

However, at Oak Springs cover and density of sagebrush were lower in grazed plots

when compared to ungrazed plots (Fig 2.2 and 2.4). Peterson et al. (2014) presented research showing the effectiveness of fall/winter grazing in reducing sagebrush cover.16

The reduction of sagebrush also leads to an increase in forbs and grasses. Similarly, other research has found that removing shrubs (sagebrush) tends to increase perennial grass and other herbaceous cover.17,18,19

In Beef Hollow native perennial grasses decreased (P=0.016) while there was an

increase in bulbous bluegrass (P=0.014); in this case, grazing is likely the factor that is

changing vegetation composition (Fig 2.6). Even though perennial grass cover was

declining, it did not result in an increase of cheatgrass cover. Research has shown that the

combination of improper grazing and/or wildfires can cause a plant community to

transition across an ecological threshold and to a new novel stable state that is dominated

by invasive annual grasses.20,21 This may have already have happened in Beef Hollow,

because the site appears to be stable. Therefore, the current grazing has not improved or

degraded the site further. Because Beef Hollow has lots of cheatgrass, however, managers should still evaluate cheatgrass cover regularly to ensure it doesn’t increase any

further. Beef Hollow has trace amounts of medusahead and jointed goatgrass, this site

should be monitored regularly to detect any increases in invasive annual grasses. 27

Wood Hollow and Beef Hollow have abundant cheatgrass cover (Fig. 2.6 and

2.9). Therefore, simply removing grazing alone would not be sufficient to return a

degraded state to higher successional state.21,22,23 Briske states that “thresholds exist at

various stages of progression which provides valuable information for defining

management and policy options regarding threshold reversibility. For example,

vegetative states that have crossed an ecological threshold but still retain a majority of

their pre-threshold species richness have a greater probability of reversal than states that

have lost most of their species and supporting ecological functions”.20 This may be true

with both Wood and Beef Hollow sites. The grazing may have been implemented

following the change in the communities. The sites may have already crossed ecological

thresholds before grazing of fuel breaks began.

Cheatgrass abundance could also be explained because of higher than normal

moisture in the spring of 2015. Casper suggested that belowground competition decreases

when there is an increase of nutrients.24 Water is a limited resource, therefore the more

abundant a resource is the less competition occurs. Precipitation in 2015 was higher than

the long-term average. In 2015 precipitation for March, April and May were 127 mm at

Camp Williams while the annual long-term average for the three months is only 104 mm.12 The additional precipitation resulted in reduced competition. This could have

resulted in increased cheatgrass abundance regardless of treatment affect. One of the limitations of this study is that the data collection only occurred in 1 year and we do not have data from other years to encompass the annual variation in cheatgrass cover.

Conclusion

In my study there were few negative changes in vegetation in the grazed fuel breaks even though they have experienced intense short-term grazing bouts annually.

Conversely, the reduction of sagebrush in Oak Springs lead to an increase of perennial

grasses which if often desirable to create heterogeneity in sagebrush communities.

Current grazing practices have not increased cheatgrass at any of the sites. This is likely

due to the grazing occurring in short intense periods. Sheep and goats were only on each

block with in the pastures for two to three days, followed by a long rest period. This is

likley the reason for stable to increasing perennial grass cover and no increase in

cheatgrass. However, grazing should be closely monitored to ensure the sustainability of

the plant communities. Invasive grasses such as medussahead and jointed goatgrass

should be monitored closely to ensure that they do not become a larger problem in the

future.

References

1. NUNAMAKER, C., M.D. LASAUX., AND G. NAKAMURA. 2007. Wildfire and fuel management. Forest Stewardship Series. University of California. http://anrcatalog.ucanr.edu/pdf/8245.pdf .Web. Feb. 2016.

2. YOUNG, J.A., R.A. EVANS, R.E. ECKERT JR, AND B.L. KAY. 1987. Cheatgrass. Rangelands 9(6):266-270.

3. PELLANT, M., 1996. Cheatgrass: the invader that won the west. US Department of the Interior, Bureau of Land Management. 22.

4. TAYLOR JR, C.A., 2006. Targeted grazing to manage fire risk. Targeted grazing: A natural approach to vegetation management and landscape enhancement. 107- 112.

5. DIAMOND, J.M., 2009. Effects of targeted grazing and prescribed burning on fire behavior and community dynamics of a cheatgrass (Bromus tectorum)- dominated landscape. Utah State University.

6. SCHMELZER, L., 2009. Reducing fuel load of key cheatgrass (Bromus tectorum) dominated range sites by the use of fall cattle grazing. Reno, NV: University of Nevada.

7. LOVINA, R., K. LAUNCHBAUGH, J. JONES, L. BABCOCK, K.R. AMBROSE, A. STEBLETON, T. BREWER, K. SANDERS, J. MINK, AND G. HYDE. 2009. Rangelands: An introduction to Idaho wild open spaces. Department of Rangeland Ecology and Management and Idaho Rangeland Resource Commission.

8. YOUNG, J.A. AND R.R. BLANK. 1995. Cheatgrass and wildfires in the intermountain west. California exotic pest plant council symposium proceedings. 1-3.

9. WHISENANT, S.G., 1990. Changing fire frequencies on Idaho's Snake River Plains: ecological and management implications. Changing fire frequencies on Idaho's Snake River Plains: ecological and management implications., (INT-276), 4-10.

10. U.S. Census, “Census of Population and Housing”. https://www.census.gov/. Retrieved June 4, 2015

11. FROST, S.M. 2015. Fire environment analysis at Army Garrison Camp Williams in relation to fire behavior potential for gauging fuel modification needs. Logan, UT, Utah State University.

12. U.S. Climate Data. “Weather History.” Your Weather Service. http://www.usclimatedata.com/ Jan.-Feb. 2016.

13. Google Earth. 2016. “Camp Williams.” 40°25'38.48"N 112° 0'49.19"W. Retrieved 9/20/2016.

14. BLM Sampling vegetation attributes: interagency technical reference. Bureau of Land Management, 1999.

15. R Core Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical computing, Vienna, Austria. URL http://www.R- project.org/.

16. PETERSEN, C.A., J.J. VILLALBA, AND F.D. PROVENZA. 2014. Influence of experience on browsing sagebrush by cattle and its impacts on plant community structure. Rangeland Ecology & Management 67(1):78-87.

17. SWANSON, S.R., J.C. SWANSON, P.J. MURPHY, J.K. MCADOO, AND B. SHULTZ. 2016. Mowing Wyoming Big Sagebrush (Artemisia tridentata ssp. wyomingensis) Cover Effects Across Northern and Central Nevada. Rangeland Ecology & Management 69(5):360-372.

18. DAVIES, K.W., J.D. BATES, AND A.M. NAFUS. 2012. Vegetation response to mowing dense mountain big sagebrush stands. Rangeland Ecology & Management 65(3):268-276.

19. BERLOW, E.L., C.M. D'ANTONIO, AND H. SWARTZ. 2003. Response of herbs to shrub removal across natural and experimental variation in soil moisture. Ecological Applications 13(5):1375-1387.

20. BRISKE, D.D., S.D. FUHLENDORF AND F.E. SMEINS. 2006. A unified framework for assessment and application of ecological thresholds. Rangeland Ecology & Management 59(3):225-236.

21. BRIGGS, J.M., A.K. KNAPP, J.M. BLAIR, J.L. HEISLER, G.A. HOCH, M.S. LETT, AND J.K. MCCARRON. 2005. An ecosystem in transition: causes and consequences of the conversion of mesic grassland to shrubland. BioScience 55(3):243-254.

22. YEO, J.J., 2005. Effects of grazing exclusion on rangeland vegetation and soils, East Central Idaho. Western North American Naturalist, 91-102.

23. LAYCOCK, W.A., 1991. Stable states and thresholds of range condition on North American rangelands: a viewpoint. Journal of Range Management 44(5):427-433.

24. CASPER, B.B. AND R.B. JACKSON. 1997. Plant competition underground. Annual Review of Ecology and Wystematics 28(1):545-570.

Figures

Figure 2.1. Camp Williams’ sheep and goat grazed fuel break pastures. Oak Spring, Wood Hollow, and Beef Hollow.

Oak Springs Brush Cover

20 B 18 16 14 12 10 8 6 % Brush Cover % Brush 4 A A A 2 A A 0 Oak brush Sagebrush Other * Grazed Ungrazed

Figure 2.2. Percent brush cover inside (grazed) and outside (ungrazed) of fuel breaks at Oak Springs in 2015 on Camp Williams. Statistical differences (p = 0.05) indicated with different letters (Other* consist of Mahonia aquifolium, Purshia tridentate, Petradoria pumila, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa, Symphoricarpos albus).

Oak Springs Cover A B 45 40 35 A 30 A B 25 A 20 A 15 A 10 A A 5 A B B % Vegetation Cover % Vegetation B 0

Grazed Ungrazed

Figure 2.3. Percent ground cover inside (grazed) and outside (ungrazed) of fuel breaks at Oak Springs in 2015 on Camp Williams. Statistical differences (p = 0.05) indicated with different letters.

Oak Springs Density

2.5 B 2 A 1.5 A 1 0.5

Average Plants/m ² Average A A A 0 Oakbrush Sagebrush Other* Grazed Ungrazed

Figure 2.4. Density of brush inside (grazed) and outside (ungrazed) fuel breaks at Oak Springs in 2015 on Camp Williams. The statistical difference (p=0.05) indicated with different letters. (Other* consist of Mahonia aquifolium, Purshia tridentate, Petradoria pumila, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa, Symphoricarpos albus).

Beef Hollow Brush Cover 12 A 10

6 A 4 A % Brush Cover Brush % A 2 A A 0 Oak brush Sagebrush Other Grazed Ungrazed

Figure 2.5. Percent brush cover inside (grazed) and outside (ungrazed) of fuel breaks at Beef Hollow in 2015 on Camp Williams Statistical differences (p = 0.05) indicated shown with different letters. (Other* consist of Mahonia aquifolium, Purshia tridentate, Petradoria pumila, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa).

Beef Hollow Cover A 45 40 A 35 B A 30 A 25 A 20 B A A 15 A 10 A A A

% Vegetation Cover % Vegetation 5 B A A 0

Grazed Ungrazed

Figure 2.6. Percent ground cover inside (grazed) and outside (ungrazed) of fuel breaks at Beef Hollow in 2015 on Camp Williams. Statistical differences (p = 0.05) indicated with different letters. (Other veg.* consist of Aegilops cylindrical, Taeniatherum caput- medusae, Opuntia polyacantha).

Beef Hollow Density 2

A 1.5 A 1 A

0.5 A

Average Plants/m ² Average A A 0 Oakbrush Sagebrush Other* Grazed Ungrazed

Figure 2.7. Density of brush inside (grazed) and outside (ungrazed) of fuel breaks at Beef Hollow in 2015 on Camp Williams. The statistical difference (p=0.05) indicated with different letters. (Other* consist of Mahonia aquifolium, Purshia tridentate, Petradoria pumila, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa).

Wood Hollow Brush Cover 10 A 9 8 7 6 5 4 3 A % Brush Cover Brush % 2 A A A 1 A 0 Oak brush Sagebrush Other* Grazed Ungrazed

Figure 2.8. Percent brush cover inside (grazed) and outside (ungrazed) of fuel breaks at Wood Hollow in 2015 on Camp Williams Statistical differences (p = 0.05) indicated with different letters. (Other* consist of Purshia tridentate, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa).

Wood Hollow A 50 A 45 40 A 35 30 A A 25 A A 20 15 A 10 A A

% Vegetation Cover % Vegetation A 5 A A A A A 0

Grazed Ungrazed

Figure 2.9. Percent ground cover inside (grazed) and outside (ungrazed) of fuel breaks at Wood Hollow in 2015 on Camp Williams. Statistical differences (p = 0.05) indicated with different letters. (Other veg.* consist of Aegilops cylindrical, Opuntia polyacantha).

Wood Hollow Density 2.5 A 2 A 1.5 1 A A 0.5 A A Average Plants/m ² Average 0 Oakbrush Sagebrush Other* Grazed Ungrazed

Figure 2.10. Density of brush inside (grazed) and outside (ungrazed) fuel breaks at Wood Hollow in 2015 on Camp Williams. The statistical difference (p=0.05) indicated shown with different letters. (Other* consist of Purshia tridentate, Chrysothamnus viscidiflorus, Gutierrezia sarothrae, Ericameria nauseosa). 40

CHAPTER 3

HOW ARE GRAZING UTILIZATION LEVELS AND FIRE BEHAVIOR RELATED?

Abstract

Wildfires in the Western United States have increased in size and frequency in recent years. Land managers have sought tools to reduce fuels and thus reduce wildfire risk. Fuel reductions also increase the ability of wildland firefighters to suppress wildfires and protect property. Grazing has been identified as a tool to potentially reduce fine fuels and thus decrease flame lengths and rate of spread. Often fine fuel treatments rely on high levels of grazing utilization (> 80%). However, high levels of utilization can lead to ecological degradation by reducing or eliminating native bunchgrasses. The objectives of this study were to determine how different levels of grazing utilization (0% [control],

30%, 50%, and 80%) relate to fuel characteristics and subsequent fire behavior. This was conducted by grazing 50 m x 20 m plots 4 reps each at 0%, 30%, 50% or 80% utilization levels (by weight). Immediately afterward, fuel measurements were collected (fuel bed depth and fuel loading). This data was then used to model flame lengths and rate of fire spread at two different windspeeds (8 kph and 40 kph) at each utilization level. Results indicate that 50% utilization and 80% utilization were no different in four of the five models for fire behavior. These models suggest that moderate grazing utilization levels allow for more sustainable fuel reduction treatments while still reducing wildfire risks.

These results ensure that fuels objectives are met while maintaining ecological integrity.

Problems and Objectives

Flame length and rate of spread are two factors that determine the intensity of a

wildfire. The more intense a wildfire is, the more damaging it may be to vegetation,

humans, and wildlife. Higher intensity wildfires can favor cheatgrass (Bromus tectorum),

which will further degrade and devalue the site. Since 1960, six of the worst wildfire

seasons have occurred from 2000 to present time.1 It has been reported that 39 of the 50

largest western wildfires from 2000 to 2009 ignited in cheatgrass-dominated sites. In that same time period, cheatgrass burned twice as often as any other vegetation type.2

Due to the growing concerns about wildfires, land managers have implemented

tools to reduce the risk of wildfire. Targeted grazing has often been suggested as a tool to

reduce fine fuel buildup. Studies generally recommend high utilization rates (80%

biomass removal) in fuel breaks to reduce fine fuel and reduce wildfire intensity allowing

for better control of wildfires.3,4,5

Overgrazing usually occurs when utilization is >65%.6 Recent fuel management

plans recommend that managers graze fine fuels at ≥80% utilization to manage wildfire intensity.3,4,5 Some range managers have become concerned that this cycle of overgrazing to reduce fuels may lead to an increase in cheatgrass abundance which will increase wildfire risks long term. The increase of cheatgrass can promote the spread of wildfires and increase the risk of more wildfires. It has been shown that 40-50% utilization is more sustainable for native vegetation whereas excessive grazing can increase cheatgrass dominance.6 It would appear that there is a conflict in management recommendations

because fuels managers are recommending 80% utilization rates while range managers 42

generally suggest taking no more than 50% in order to maintain the health of perennial grasses. It is clear that rangeland and fuels managers need to understand how utilization levels influence fire behavior.

The objective of my research was to examine the relationship of utilization rates on fire behavior. To do this, I evaluated fire behavior at different forage utilization rates

(0%, 30%, 50% and 80%) to better understand this relationship and to determine if land managers can use lower utilization rates while still reducing wildfire risk.

Methods

My research was conducted south of Richmond, Utah, in September and October

2015 at the Utah State University Agriculture Experiment Station North Farm (elevation:

1511 m, precipitation: 525.8 mm). The site was formerly enrolled in the Conservation

Reserve Program (CRP). The site is dominated by smooth brome grass (Bromus inermis).

In September of 2015, there was 4890 kg/ha of grass at the study site. Utilization treatments included 0 (control), 30, 50, and 80% utilization (by weight) were replicated 4 times for a total of 16 pastures that were 50m x 20 m. The pastures in each replication were randomized at different utilization levels although control plots were placed at the end of each block due to logistical constraints in moving cattle between pastures and marinating the integrity of the control pastures (Table 3.1).

I clipped biomass in each pasture before cattle were introduced. Stocking rates for

the 30% utilization treatments averaged 5.2 AUM/ha. Average stocking rate for the 50%

treatments was 10.08 AUM/ha. The stocking rate average for the 80% utilization

treatments was 13.92 AUM/ha. I monitored utilization rates in each pasture daily by 43

clipping 5 0.10 m2 frames. After prescribed utilization levels were met, cattle were

removed, and fuel bed depth and fuel load data were collected. Fuel bed and fuel load

data were collected in 5 0.5m² frames randomly located in each pasture. Fuel bed depth

was measured by taking height measurements of vegetation at the four corners of the

quadrat to get average fuelbed depth. Fuel loads were measured by clipping and

collecting the remaining fuel in each 0.5m² frame (Table 3.2). Clippings were separated

into live and dead foliage by fuel class and bagged. The samples were dried (48 hours at

65° C) and weighed.

BeHavePlus5 was used to model fire behavior across the different utilization

levels. BeHave Plus5 is a wildfire behavior modeling program.7 Multiple fuel and

weather inputs are used to generate the fire behavior models (Table 3.3). The model outputs were flame length, the rate of spread, and the chance of ignition.

Fire behavior was modeled under 2 scenarios (8 kph and 40 kph). Outputs were then entered into Rstudio to be analyzed. An lmerTest was used to detect differences at a

P=0.05 level between each treatment for flame length, the rate of spread, and percent chance of ignition.

Results

Fuel bed depth (height) ranged from, 73 cm in the control pastures to 18 cm in the

80% utilization pastures. Fuel loads ranged (weight) from 4890 kg/ha in control pastures to 1500 kg/ha, respectively.

Flame Length 44

Modelled flame lengths at 8 kph wind speed were (control) were 1.99 m while

30%, 50%, and 80% utilization were 1.19 m, 0.89 m, and 0.58 m respectively. Modeled

flame lengths from control plots were longer that all other treatments (P=0.0001). At 30% utilization, the modeled flame lengths were taller than flame lengths from 80% utilization plots (P=0.0006) and 50% utilization flame lengths were longer when compared to 80% utilization (P=0.0336). However, 50% flame lengths were not different from flame lengths from 30% utilization plots (P=0.054) (Table 3.1).

Flame length at 40 kph were 2.95 m for control plots. Treatments of 30%, 50%, and 80% modeled flame lengths at 1.52 m, 1.01 m and 0.58 m respectfully. All utilization treatments (30% 50% and 80%) had shorter flame lengths when compared to control plots. When comparing the treatments, the only difference was between the 80% and

30% treatments (P=0.0058). No difference in flame lengths at 40 kph were found when comparing 30% and 50% (P=0.126) or 50% and 80% (P=0.226) (Table 3.1).

Rate of Spread

Pastures with no grazing had an average rate of spread of 0.92 kph at wind speeds of 8 kph. While treatments levels of 30%, 50%, and 80% utilization displayed rates of spread of 0.46 kph, 0.31 kph, and 0.18 kph at 8 kph. Control plots showed differences compared to all other treatments (P=0.0019, P=0.0002, and P=0.0001, respectively). The rate of spread was faster at 30% utilization when compared to 80% utilization. (P=0.035).

However, the rate of spread was similar when comparing 30% utilization 50% utilization

(P=0.317), also the rate of spread was similar when comparing 50% utilization to 80% utilization (P=0.469) (Table 3.2). 45

The modeled rate of spread measured at 40 kph for control plots was 2.15 kph, while treatments of 30%, 50%, and 80% utilization were 0.76 kph, 0.39 kph, and 0.18

kph respectively. Similar to 8 kph scenario the control plots were different when

compared to treatments. 30% utilization had greater rate of spread compared to 80%

utilization (P=0.03) with no difference between 30% and 50% (P=0.19) and 50% and

80% utilization rates were similar (P=0.63) (Table 3.2).

Chance of ignition across all wind speeds were the same (16%). Indicating no

difference between treatments, having a P=1.00.

Discussion

All grazing treatments reduced flame lengths in the kph scenario compared to

control plots. Even though there are some statistical difference in flame lengths between

the treatments, the most important consideration is, whether grazing reduces flame

lengths 1.22 m. This is an important threshold for wildland fire fighting operations, when

flame lengths are below 1.22 m hand crews can use direct firefighting techniques. If

flame lengths exceed 1.22 m, then hand crews have to use indirect methods of

firefighting making fire control more difficult. Under the 8 kph, wind speed scenarios of

all grazing treatments (30, 50 and 80%) flame lengths were less than 1.22m (1.19 m, 0.89

m, and 0.58 m respectively) making direct fire suppression activities possible (Table 3.1).

It should be noted that utilization levels of 50% and 80% reduced flame lengths to less

than one meter.

When evaluating the 40 kph wind speed scenarios, targeted grazing only reduced

flame lengths below 1.22 m in the 50% and 80% utilization treatments. However, flame 46

lengths at 50 and 80% utilization grazing rates were not different at 40 kph wind speed

(Table 3.1).

Therefore, results from the model suggest that there is no advantage of utilization

rates over 50% since higher utilization rates will not significantly reduce flame lengths.

Managers may be able to reduce grazing intensity while maintaining fire suppression.

This could help managers use a moderate grazing intensity level which will allow a better

chance for perennial grasses to survive while reducing wildfire fuels.

The rate of spread at 8 kph wind speed in the control was 0.92 km/hr. All grazing treatments reduced fire rate of spread demonstrating that grazing will influence fire behavior even at light utilization rates. However, there was no significant difference in

rates of spread between 50 and 80% utilization rates. We see a similar trend with the 40

kph wind speed scenarios showing no difference in the rate of spread between 50 and

80% utilization rates. These results suggest that increasing utilization above 50% utilization will reduce the rate of spread significantly (Table 3.2).

While other researchers have suggested utilization rates of ~80% none of them have reported the differences in fire behavior under different utilization scenarios. For example, Diamond conducted a spring grazing trial, while Schmelzer had a similar study when looking at cheatgrass response to heavy fall grazing. Both research projects evaluated only an 80% utilization rate in cheatgrass-dominated sites and its effects on wildfire control.3,5 However, neither of these studies examined different rates of

utilization and their effect on wildfires. The level of utilization in these studies could be

harmful to native perennial vegetation on arid rangelands. Strand et al. found that 47

utilization rates >50%, especially in spring, can cause a decrease in native perennial

grasses and increases invasive grasses.8

A noticeable difference between the Richmond study site and all other grazing

fuel treatment studies is that the Richmond site is dominated by smooth brome (Bromus inermis) with production around 4898 kg/ha, while all other grazing fuel treatment studies tend to be in more arid less productive rangelands. These rangelands tend to have a mixture of cheatgrass, native grasses, and sagebrush. Fine fuel production in the

Richmond study is higher than what would be expected in a typical sagebrush rangeland setting. For example, the 80% utilization treatment had 1542 kg/ha fuel after grazing was complete, which is more fuel than many sagebrush sites would produce in a year. It has been reported that cheatgrass dominated sites (without treatment) consist of an average of

500-600 kg/ha and can range from 30 kg to >1500 kg/ha.9 Fuel continuity in Richmond is also higher than what would be found on a typical rangeland setting because smooth brome is a sod forming grass. (See Appendix for all tables).

Conclusion

My research demonstrates that under the conditions of our model that there are no real advantages to grazing utilization levels exceeding 50%. Neither flame lengths or rate of spread were lower under the 80% unitization when compared with 50% utilization.

Therefore, there is little justification to graze so intensely because fire behavior does not significantly change with utilization above 50%. Although my results show a reduction in wildfire behavior, more research needs to be conducted in native rangelands to validate 48

my findings. Future research should also attempt to use actual fire behavior to evaluate

differences in utilization rates.

References

1. GORTE, R. AND Headwater Economics. 2013. The rising cost of wildfire protection. Headwaters Economics.

2. BALCH, J.K., B.A. BRADLEY, C.M. D'ANTONIO, AND J. GOMEZ‐DANS. 2013. Introduced annual grass increases regional fire activity across the arid western USA (1980–2009). Global Change Biology 19(1):173-183.

3. DIAMOND, J.M., 2009. Effects of targeted grazing and prescribed burning on fire behavior and community dynamics of a cheatgrass (Bromus tectorum)- dominated landscape. Utah State University.

4. TAYLOR JR, C.A., 2006. Targeted grazing to manage fire risk. Targeted grazing: A natural approach to vegetation management and landscape enhancement, 107- 112.

5. SCHMELZER, L., 2009. Reducing fuel load of key cheatgrass (Bromus tectorum L.) dominated range sites by the use of fall cattle grazing. University of Nevada, Reno.

6. LOVINA, R., K. LAUNCHBAUGH, J. JONES, L. BABCOCK, K.R. AMBROSE, A. STEBLETON, T. BREWER, K. SANDERS, J. MINK, AND G. HYDE. 2009. Rangelands: An introduction to Idaho’s wild open spaces. Department of rangeland ecology and management, University of Idaho.

7. HEINSCH, F.A. AND P.L. ANDREWS. 2010. BehavePlus fire modeling system, version 5.0: design and features.

8. STRAND, E.K., K.L. LAUNCHBAUGH, R.F. LIMB, AND L.A. TORELL. 2014. Livestock grazing effects on fuel loads for wildland fire in sagebrush dominated ecosystems. Journal of Rangeland Applications 1:35-57.

9. URESK, D.W., J.F. CLINE AND W.H. RICKARD. 1979. Growth rates of a cheatgrass community and some associated factors. Journal of Range Management 32(3):168-170.

10. Admin. (2005, November 07). Unit 5: Fuel Moisture. Free Online Course Materials USU OpenCourseWare Web site: http://ocw.usu.edu/Forest__Range__and_Wildlife_Sciences/Wildland_Fire_Mana gement_and_Planning/Unit_5__Fuel_Moisture_8.html. Retrieved January 07, 2011

Tables and Figures

Table 3.1. Plot layout for grazing utilization trial at Richmond, UT.

Richmond Utah Grazing Plot Layout Control 80% 50% 30% Control 50% 30% 80% Control 30% 80% 50% Control 80% 50% 30%

Table 3.2. Richmond fuel biomass averages collected post-grazing.

Richmond Fuel Average Utilization Rate Fuel Load (kg/ha) Fuel Bed Depth (m) 0% (control) 4890 0.72 30% 3140 0.41 50% 2350 0.26 80% 1500 0.18

Table 3.3. BeHavePlus5 inputs and data source for inputs. Dead Fuel Moisture of Extinction received from Utah State University/Admin.11

BeHavePlus 5 Inputs: Obtained: Fuel Model Type Dyna Due to live fuel component mic 1-hour Fuel Load Data Collected 10-hour Fuel Load NA 100-hour Fuel Load NA Live Herbaceous Fuel Data Collected Load Live Woody Fuel Load NA 1-hr SA/V 1000 Average for Course Grass in BeHavePlus5 Live Herbaceous SA/V 1500 Average Grass in BeHavePlus5 Live Woody SA/V 0 Fuel Bed Depth Data Collected Dead Fuel Moisture of 25% (USU Admin 2007)10 Extinction Dead Fuel Heat Content 8000 Average given in BeHavePlus5 btu Live Fuel Heat Content 8000 Average given in BeHavePlus5 btu Dead Fuel Moisture 15% Live Fuel Moisture 41% Midflame Wind Speed Randomly Generated Air Temperature 32.2º Average in Richmond, UT in August C Fuel Shading from the 0 Sun Slope 0

Flame Length (m) 3.50

3.00

2.50

2.00

1.50

1.00 Flame (m) Length 0.50

0.00 0 30 50 80 Utilization %

8.05 kph wind speed 40.23 kph wind speed 1.22 m Flame Length

Figure 3.1. Modeled flame length (meters) with two different wind speed scenarios at 0%, 30%, 50% and 80% utilization rates. The dotted line represents a 1.22m flame length (4ft) where direct firefighting methods are no longer possible.

Rate of Spread kph 3.00

2.50

2.00

1.50

1.00

Rate of Spread kph Spread of Rate 0.50

0.00 0 30 50 80 Utilization %

8.05 kph wind speed 40.23 kph wind speed

Figure 3.2. The Modeled rate of spread at kilometers per hour with two different wind speed scenarios at 0%, 30%, 50%, and 80% utilization rates. 53

CHAPTER 4

CONCLUSION AND DISCUSSION

The purpose of my research was to evaluate targeted grazing as a fuels treatment and to explore the relationship between grazing utilization rates and fire behavior. When evaluating the fuel breaks for grazing impacts, I found few differences between grazed and ungrazed portions of the pastures. Oak Springs exhibited the most changes between grazed and ungrazed plots. Targeted grazing decreased sagebrush and increased herbaceous cover. The most significant finding at Beef Hollow was an increase in bulbous bluegrass while having a decrease in native bunchgrass (50% decrease) compared to the grazed site. The site also was the only site with trace amounts of medusahead, along with jointed goatgrass, but it was present in referenced and grazed pastures. In Wood Hollow, there were no significant changes in vegetation cover.

However, Wood Hollow had a high cheatgrass cover above 35%, it was equally abundant in reference and grazed plots. Wood Hollow may have already crossed an ecological threshold and will likely increase cheatgrass cover with additional disturbances such as fire.

My results also indicated that targeted grazing caused no significant increase of cheatgrass. This may be due to the short intense grazing followed by a year rest before next grazing in the pastures.

Targeted grazing in Richmond, showed that all treatments were different from control in both flame length and rate of spread when modeled at two different wind speed scenarios. My research suggested that 50% utilization rates were not different from 80% 54 utilization rates in terms of flame lengths or rate of spread and both levels are well below

1.22 (4ft) level that allows a more direct attack on wildfires. These results suggest that fire managers can use more conservative utilization rates while reducing rate of spread and flame length thus reducing fire intensity and allowing for better control of wildfires.

Conclusion

Grazing as a tool for fuel management when implemented with proper intensity and timing leave rangelands less susceptible to high-intensity wildfires. Based on my research, managers can lower utilization levels and still reduce the flame length and rate of spread. In this study targeted grazing didn’t cause widespread degradation.

APPENDIX

Table A.1. Cover percent and p-values of herbaceous plants along with bare ground and litter in Oak Springs. Showing between grazed (treatment) and ungrazed (reference) averages. (Other brush consists of Oregon grape, antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush, and common snowberry). Asterisks (*) shows difference at P=.05 level.

Oak Springs Cover Grazed Ungrazed Means SE± Means SE± P-Value Oakbrush 1.00 0.77 1.00 0.59 0.6572 Sagebrush 0.00 0.02 16.00 2.58 0.0001* Other brush 1.00 0.34 2.00 0.69 0.6107 Cheatgrass 15.00 4.23 10.00 3.19 0.1322 Bulbous bluegrass 1.00 0.61 0.00 0.00 0.0492* Native grasses 38.00 5.55 26.00 4.56 0.5759 Introduced forbs 16.00 1.29 1.00 0.44 0.0001* Native forbs 5.00 1.03 1.00 0.63 0.0198* Bare 22.00 3.15 38.00 5.05 0.0180* Litter 3.00 1.48 24.00 2.83 0.0001*

Table A.2. Cover percent and p-values of herbaceous plants along with bare ground and litter in Beef Hollow. Showing between grazed (treatment) and ungrazed (reference) averages. (Other veg. consists of jointed goatgrass, medusahead grass, pricklypear cactus. Other brush consists of Oregon grape, antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush). Asterisks (*) shows difference at P=.05 level.

Beef Hollow Cover Grazed Ungrazed Means SE± Means SE± P-Value Oakbrush 1.77 0.78 5.56 4.64 0.632 Sagebrush 0.89 0.59 2.73 1.17 0.225 Other brush 0.67 0.35 1.25 0.56 0.879 Cheatgrass 37.29 6.26 22.71 5.45 0.185 Bulbous bluegrass 10.21 3.77 0.83 0.55 0.014* Native grasses 12.71 1.89 28.13 4.61 0.016* Introduced forb 17.92 6.34 5.42 1.99 0.062 Native forb 3.54 1.42 3.33 1.70 0.739 Bare 15.00 4.75 25.83 7.31 0.135 Litter 1.88 0.66 12.08 4.82 0.038* Other 5.21 3.98 1.67 0.70 0.863

Table A.3. Cover percent and p-values of herbaceous plants along with bare ground and litter in Wood Hollow. Showing between grazed (treatment) and ungrazed (reference) averages. (Other veg. consists of jointed goatgrass, pricklypear cactus. Other brush consists of antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush). Asterisks (*) shows difference at P=.05 level.

Wood Hollow Cover Grazed Ungrazed Means SE± Means SE± P-Value Oakbrush 2.0 0.86 6.00 3.10 0.251 Sagebrush 1.0 0.57 1.00 0.50 0.819 Other brush 0.0 0.05 1.00 0.26 0.083 Cheatgrass 41.00 6.73 37.00 7.08 0.677 Bulbous bluegrass 2.00 1.37 0.00 0.21 0.181 Native grasses 16.00 4.54 18.00 7.27 0.785 Introduced forb 29.00 6.95 22.00 5.01 0.536 Native forb 3.00 2.69 2.00 0.93 0.670 Bare 8.00 2.31 16.00 5.05 0.284 Litter 0.00 0.42 2.00 1.22 0.113 Other 0.00 0.35 1.00 0.46 0.351

Table A.4. Density and p-values of brush/shrubs at Oak Springs. Other brush consists of (Oregon grape, antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush, and common snowberry). Asterisks (*) shows difference at P=.05 level.

Oak Springs Density Grazed Ungrazed Means SE± Means SE± P-Value

Oak brush 0.75 0.62 0.88 0.88 0.872

Sagebrush 3.88 3.73 58.25 10.24 .0001*

Other brush 32.25 11.94 27.13 7.77 0.646

Table A.5. Density and p-values of brush/shrubs at Beef Hollow. Other brush consists of (Oregon grape, antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush). Asterisks (*) shows difference at P=.05 level.

Beef Hollow Density Grazed Ungrazed Means SE± Means SE± P-Value Oakbrush 8.75 4.09 19.75 12.85 0.783

Sagebrush 2.13 1.11 6.50 2.44 0.144

Other brush 19.50 7.46 31.63 14.69 0.826

Table A.6. Density and p-values of brush/shrubs at Wood Hollow. Other brush consists of (Antelope bitterbrush, yellow rabbitbrush, broom snakeweed, rubber rabbitbrush). Asterisks (*) shows difference at P=.05 level.

Woods Hollow Density Grazed Ungrazed Means SE± Means SE± P-Value Oakbrush 16.50 9.34 44.50 23.97 0.344 Sagebrush 8.63 5.20 5.88 2.45 0.838 Otherbrush 6.38 3.28 31.63 16.41 0.353

Table A.7. Flame lengths at 8.05 kph in meters. Showing the control and the three different treatment means, standard error, and p-value. Asterisks (*) shows no difference between treatments at P=0.05.

8.05kph Flame Length (m) Means SE± Control (0%) 1.990 0.098 30% 1.190 0.109 50% 0.890 0.015 80% 0.580 0.064

Treatments P-Value 0%-30% 0.0001 0%-50% <.0001 0%-80% <.0001 30%-50% 0.0546* 30%-80% 0.0006 50%-80% 0.0336

Table A.8. Flame lengths at 40.23 kph in meters. Showing the control and the three different treatment means, standard error, and p-value. Asterisks (*) shows no difference between treatments at P=0.05.

40.23kph Flame Length (m) Means SE± Control (0%) 2.950 0.260 30% 1.520 0.147 50% 1.010 0.058 80% 0.580 0.064

Treatments P-Value 0%-30% 0.0003 0%-50% <.0001 0%-80% <.0001 30%-50% 0.1261* 30%-80% 0.0058 50%-80% 0.2265*

Table A.9. Rate of spread in kph with 8.05kph mid-flame wind speeds. Showing the control and the three different treatment means, standard error, and p-value. Asterisks (*) shows no difference between treatments at P=0.05.

8.05kph Rate of Spread (kph) Means SE± Control (0%) 0.915 0.044 30% 0.458 0.119 50% 0.305 0.027 80% 0.175 0.031

Treatments P-Value 0%-30% 0.0019 0%-50% 0.0002 0%-80% <.0001 30%-50% 0.3173* 30%-80% 0.0354 50%-80% 0.4690*

Table A.10. Rate of spread in kph with 40.23kph mid-flame wind speeds. Showing the control and the three different treatment means, standard error, and p-value. Asterisks (*) shows no difference between treatments at P=0.05.

40.23kph Rate of Spread (kph) Means SE± Control (0%) 2.150 0.241 30% 0.760 0.142 50% 0.385 0.019 80% 0.175 0.031

Treatments P-Value 0%-30% 0.0001 0%-50% <.0001 0%-80% <.0001 30%-50% 0.1964* 30%-80% 0.0327 50%-80% 0.6360*