Distribution and Associated Arthropods of Native and Non-Native Thistles in the Valles Caldera National Preserve, New Mexico, USA

Distribution and Associated Arthropods of Native and Non-Native Thistles in The Valles Caldera National Preserve, New Mexico, USA

by Neil Estes, B.Sc. A Thesis In Wildlife, Aquatic and Wildland Science Management Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE

Robin Verble, Ph.D. Chair of Committee

Warren Conway, Ph.D. Robert Cox, Ph.D. Robert Parmenter, Ph.D.

Mark Sheridan, Ph.D. Dean of the Graduate School

May 2018

ABSTRACT

Non-native plant species represent an important threat to rangeland health and productivity. In New Mexico highland range, four non-native species of interest, Cirsium vulgare, C. arvense, Carduus nutans, and Bromus tectorum, are irregularly and patchily distributed across a diverse and disturbed landscape.

In this study I surveyed the Valles Caldera National Preserve (VCNP) for target non-native species and mapped their distribution and abundance.

I measured trends in non-native species abundance and distribution using road surveys, chance encounters, and targeted search efforts. I then correlated non-native species distribution to fire treatment, burn severity, aspect, and logging history with none being significant. Non-native plant species occurred more frequently at burned sites then at unburned sites, with high severity sites having the highest occurrence of invasive plants. Furthermore, areas of dense forest that are being thinned to mitigate fire danger may exhibit the same non- native species response as areas impacted by high severity burns.

Arthropod community interactions with native and non-native thistle species are also an important component of rangeland health and productivity.

Arthropod taxonomic richness and abundance were compared between one non- native species of thistle, Cirsium vulgare, and one native species of thistle, C. pallidum. I collected 22 C. vulgare thistle plants and 24 C. pallidum thistle plants and then sampled and identified the insect families and functional groups present on those plants. I found that spiders and moths/butterflies showed differences between thistle species. Beetles were more abundant on the native thistle. We also grouped taxa by functional feeding group using readily available

ii Texas Tech University, Neil Estes, May 2018 life history information. Taxonomic richness differed between thistle species, with native thistles having significantly more taxa. Predators and pollinators were more abundant on native thistles.

These results will inform land management decisions regarding non- native species occurrence in areas of high disturbance like those impacted by restoration efforts and or wildland fire. While it is difficult to predict the occurrence and/or frequency of disturbance in many cases, understanding post- disturbance trends is vital when managing land for productivity long term. As wildland fire continues to increase in frequency and intensity in the southwestern

United States, management issues such as the ones being addressed on VCNP

(e.g., biodiversity preservation and biological invasions) will become more commonplace and widespread. Thistle invasion has the potential to comprise long-term rangeland health and productivity, and insect communities that occur on these thistles may represent case studies for more wide-scale responses to plant invasions. In addition, local insect communities provide a cascade of benefits to their ecosystem (e.g., pollination, prey, decomposition, predation, parasitism, etc..) and invasion-driven changes in abundance and diversity may compromise ecosystem health.

iii Texas Tech University, Neil Estes, May 2018

Table of Contents

Abstract ...... i

List of Tables ...... v

List of Figures ...... vi

1 Literature Review

Land Use History of the VCNP ...... 1

Fire History in the VCNP ...... 3

Non-Native Plants of Interest ...... 5

Effects of Non-Native Plants on Insect Communities ...... 12

Literature Cited ...... 19

2 Thistle distribution and occurrence in the VCNP

Abstract ...... 26

Introduction ...... 27

Materials and Methods ...... 28

Results ...... 31

Discussion ...... 32

Literature Cited ...... 36

Tables ...... 38

Figures ...... 39

3 Arthropod communities on native and non-native thistle in the VCNP

Abstract ...... 42

Introduction ...... 43

Materials and Methods ...... 44

iv Texas Tech University, Neil Estes, May 2018

Results ...... 46

Discussion ...... 47

Literature Cited ...... 50

Tables ...... 52

Figures ...... 56

4 Management Recommendations

Discussion ...... 61

Literature Cited ...... 64

v Texas Tech University, Neil Estes, May 2018

List of Tables

2.1 Contingency analysis of thistle cluster size (small, medium, large, extra large) by aspect (north, south, east, west). P < 0.05 was considered significant………………………………….……..…40

2.2 Contingency analysis of thistle cluster size by fire severity. P < 0.05 was considered significant………………………………………...…41

2.3 Contingency analysis of thistle occurrence by aspect. P < 0.05 was considered significant…………………………….…….…….…42

2.4 Contingency analysis of thistle occurrence by fire severity. P < 0.05 was considered significant…………………………………..…….…43

3.1 One-way ANOVA of arthropods by thistle type. DF = degrees of freedom. SS = sum of squares. MS = mean square. *=significant (p < 0.05). Abundance is the number of individuals per plant. Arthropod richness is the number of orders present on each individual plant……………………………….…..…58

vi Texas Tech University, Neil Estes, May 2018

List of Figures

2.1 Location and type of generated points on the VCNP for purpose of thistle sampling……………………………………………………………….……44

2.2 Locations of found focal thistle by species within the perimeter of VCNP. …………………………………………………………………..………….…45

2.3 Non-metric multidimensional scaling ordination of focal thistle species (C. nutans, musk thistle; C. vulgare, bull thistle; and C. arsense, Canada thistle) occurrence with fitted treatment groups of burn severity (high, low, unchanged) and aspect (north, south east, west). Stress = 0.092, Dimensions = 2. ……….…46

3.1 Araneae relative abundance per plant (Y) marginally differs between native and non-native thistle (p = 0.05). Cirsium pallidum = native thistle. Cirsium vulgare = non-native thistle. Bars represent standard error……….………………………….….…62

3.2 Lepidoptera relative abundance (Y) is marginally different between native and non-native thistle (p = 0.05). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error. …………………………………….…63

3.3 Arthropod relative abundance (Y) is marginally different between native and non-native thistle (p = 0.07). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error. ………………………………….……64

3.4 Arthropod richness (i.e., the number of individual orders found on each plant; Y) is marginally different between native and non-native thistle (p = 0.004). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error. …………………………………………….……………65

3.5 Coleoptera relative abundance (Y) differs between native and non-native thistle (p = 0.04). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error……………………………………………………………………………………..…66

3.6 Hymenoptera relative abundance (Y) differs between native and non-native thistle (p = 0.04). Cirsium pallidum = native thistle.

vii Texas Tech University, Neil Estes, May 2018

Cirsium vulgate = non-native thistle. Bars represent standard error…………………………………………………………………….……………….…67

3.7 Predator relative abundance (Y) differs between native and non-native thistle (p = 0.04). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error…………………………………………………………………….….…68

3.8 Plant dweller relative abundance (Y) differs between native and non-native thistle (p = 0.09). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error………………………………………………………….………………69

3.9 Pollinator relative abundance (Y) differs between native and non-native thistle (p = 0.056). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error……………………………………………………………………….…70

viii Texas Tech University, Neil Estes, May 2018

CHAPTER 1

LITERATURE REVIEW

Land Use History of the Valles Caldera National Preserve

Past land use is vital to understanding current community composition and structure (Harding et al. 1998). The Valles Caldera National Preserve

(VCNP) has a very extensive land use history that began with Native American colonization. The first people to provide archeological evidence of their presence in the area were hunters of large game animals that moved in and out of the area periodically between 9500 and 5500 B.C.E. (Anschuetz et al. 2007). Many native cultures used the area over the next several millennia, though it is doubtful the caldera was ever used as a permanent settlement by native people during these earlier times (Martin 2003). It is more likely that temporary camps were established along the edges of and in valleys leading to the Caldera, allowing hunter gatherer societies access to the higher elevations during the warm seasons

(Martin 2003). Permanent settlement was not established in the area until the year 500 when the people of the Ancestral Pueblo culture lived year-round in the adjacent foothills and plains (Martin 2003) and commonly travelled into the surrounding high country to take advantage of fertile soils for their crops (Martin

2003).

The first European accounts of the area began when Spanish expeditions explored the areas surrounding the Valles Caldera beginning in the middle of the sixteenth century (Anschuetz et al. 2007). The first account of the Caldera by

European explorers was written by a professional miner named Juan de Orñate.

1 Texas Tech University, Neil Estes, May 2018

Because of his expertise as a miner, Orñate was given a contract to colonize New

Mexico in 1598 and established the first permanent modern settlement

(Anschuetz et al. 2007). As a mining expert, Orñate explored many different parts of New Mexico and passed through the Caldera on his way from San Juan

Pueblo to the other Jémez villages (Anschuetz et al. 2007). The Spanish occupied

New Mexico until 1846 when U.S. troops invaded from the east. New Mexico became a U.S. territory in 1850 after the signing of the Treaty of Guadalupe

Hidalgo (Anschuetz et al. 2007).

By 1851 the rangelands surrounding VCNP had been depleted by drought and overgrazing. The U.S. Army was actively fighting Native Americans in the area from the 1850’s to the 1880’s (Anschuetz et al. 2007). As a result, much of the Jémez Mountains were converted into private land grants. Baca Location No.

1 was established as part of a land exchange in 1860 between the Baca family and the U.S. government. The Baca Location included Valle Grande, Valle San

Antonio, the Valle Santa Rose, and Redondo Creek (Anschuetz et al. 2007). The property became known as the Baca Ranch. Ownership was then transferred to

Frank Bond in 1926 (Muldavin and Tonne 2003). Bond was a sheep rancher and after acquiring the land, he increased the number of sheep and horses grazing on the property. In the 1930’s the New Mexico Timber Company took advantage of an old timber lease and began major timber harvests within the Caldera

(Muldavin and Tonne 2003). In the 1940’s, Bond added cattle to his operations in the Caldera because of dropping wool prices. By the mid 20th century approximately 12,000 cattle and 30,000 sheep grazed grasslands within the

Caldera. James Dunigan purchased the Baca Ranch in 1963 and began grazing

2 Texas Tech University, Neil Estes, May 2018 his own cattle on the property when grazing permits expired in 1965. Damage from overgrazing was extensive by this time. Dunigan introduced a rotational grazing system into the Caldera in an attempt to reduce the century long accumulation of damage (Anschuetz et al. 2007). The Valles Caldera National

Preserve (VCNP) was established when The Valles Caldera Preservation Act of

2000 was signed by President Bill Clinton on 25 July 2000 (Little et al. 2005).

The Dunigan family sold the 38,445 hectare property along with 7/8ths of the geothermal mineral estate for $101 million (Anschuetz and Merlon 2007). The management of the property then passed to a private trust until 30 September

2015 when the National Park Service officially took over management responsibilities.

Fire History of the Valles Caldera National Preserve

Fire is a fundamental process in maintaining the ecological structure and function of high elevation ecosystems in the Southern Rocky Mountains (Weaver

1951). Fire is a keystone ecological process, thus it is important to understand historic fire regimes as well as modern fire history, as these directly influence decisions about management of high elevation ecosystems such as the VCNP

(Allen 1994). Historic fire regimes in high elevation ponderosa pine systems were frequent, low-intensity surface fires. Prior to modern settlement, these forest systems exhibited an open structure because of the frequent surface fires that promoted grasses and limited tree encroachment (Touchan et al. 1996).

Montane grassland habitats are generally rare and only exist in a few areas in

3 Texas Tech University, Neil Estes, May 2018

New Mexico and in some small patches in Colorado, thus their conservation is of significant interest.

Historically, wildland fires did not rely on human ignition. Instead, there are many forms of natural ignition including lightning, sparks from falling rocks, volcanic activity, and even spontaneous combustion of organic material (Agee

1993). Lightning however, contributes to the ignition of over 6,000 fires annually in the United States (Smith 2000) and is the most common form of natural ignition. In the southwestern parts of the United States, the high frequency of lightning means that natural ignitions are still very important in shaping the ecology (Allen 2002). Southwestern ponderosa pine forests were shaped through time by stochastic and deterministic processes such as frequent surface fires, episodic regeneration, insect infestations, and regional climate events such as drought (Dahm and Geils 1997, Allen and Breshears 1998, Kaufmann et al. 1998,

Swetnam and Betancourt 1998, Allen 2002). The historic fire regime changed abruptly when Euro-Americans settlers began moving west and arrived in the area in the late 1800s. In addition to the historical account, this change is recorded in charcoal and tree-ring records (Swetnam and Baisan 1996, Allen et al. 2008). Livestock was also brought in to take advantage of abundant grasses in numbers greater then the land could sustain, reducing the fuel and slowing the fire regime (Touchan et al. 1996). Fires were viewed by the settlers as dangerous to life and property, so any surface fires were suppressed further breaking the cycle. Roads and railroads were built, disrupting the continuity of fuels and making it more difficult for fire to propagate. Logging was widespread and unchecked. All of these factors contributed to a significant shift in the ecology of

4 Texas Tech University, Neil Estes, May 2018 the area; the modern fire regime is dramatically different from what it was only centuries earlier (Allen 2002). Greater fuel loads, habitat fragmentation, fire suppression and climate change have all contributed to a low frequency, high- intensity fire regime (Covington and Moore 1994).

In recent years, several large wildfires have occurred on or burned into the

VCNP. In 2011 the Las Conchas wildfire spread across the eastern borders of the

VCNP. The wildfire burned 63,131 ha, including about 12,140 ha within the

VCNP. In 2013 the Thompson Ridge wildfire burned about 9,712 ha. This wildfire’s perimeter was completely within the VCNP. These wildfires have helped shape the land management techniques currently being utilized on the

VCNP.

Nonnative Plants of Interest

Fire suppression and poor land management over the last century have altered the structure, composition, and ecology of western rangelands (Keane and

Crawley 2002). Many of the modern rangeland restoration techniques, (e.g., thinning and burning), have the potential to facilitate invasion by nonnative plants, which could compromise restoration efforts (D’Antonio and Meyerson

2002). Roads are another route by which invasion can occur as they can act as a conduit for the spread of nonnative plants into neighboring habitats, possibly facilitating distribution of seeds by wind, increased animal traffic or increased vehicle traffic (Gelbard and Belnap 2003, Christen and Matlack 2009, Flory and

Clay 2009). The invasion of nonnative plants into new habitats in which they have not coevolved with the native biota is a major threat to biodiversity and

5 Texas Tech University, Neil Estes, May 2018 ecosystem structure and function (Mack et al. 2000, Vitousek et al. 1997). The

Valles Caldera National Preserve has engaged in an effort to restore its rangelands and forests through thinning and prescribed burning programs. Also, as a result of the logging and ranching activities in the past within the VCNP, an extensive road system of about 1931 km exists on the property today. Because of the previously mentioned risks of invasion within the VCNP, three potentially invasive thistles will be mapped and studied in this project.

C. vulgare (bull thistle) has long been naturalized in North America, probably introduced by impure seed or by some other accidental means

(WSNWCB 2016). Native to Europe, Asia and Northern Africa, it currently can be found throughout the United States north and Canada (WSNWCB 2016). Very little specific information is known about the climatic requirements of C. vulgare as it is broadly distributed and seems adapted to a wide range of climates

(WSNWCB 2016). This plant prefers heavier soils with high nitrogen contents

(WSNWCB 2016). This preference for high soil fertility has made the plant a problem in areas where fertilizer is used (i.e., agricultural fields; WSNWCB

2016). Improved pastures or cultivated areas allow the plant to establish and spread, especially if the soil is bare towards the end of summer (WSNWCB 2016).

C. vulgare can often be seen in pastures, along roadsides, in ditches, hayfields, prairies and areas that have been logged (Andersen and Louda 2006). It does well in areas of high disturbance. Other common names include common thistle and spear thistle (USDA 2016).

C. vulgare is a biennial herbaceous plant growing from 1 – 2 m tall in an upright branched system (WSNWCB 2016). Stems are very prickly and winged at

6 Texas Tech University, Neil Estes, May 2018 the leaf bases with an alternate leaf arrangement. Basal leaves can be up to a foot long and are deeply lobed and spine tipped (WSNWCB 2016). Stem leaves are smaller at about 6 – 20 cm long. The upper leaf surface is hairy while the bottom is covered in white wooly hairs (WSNWCB 2016). C. vulgare produces multiple flowers, each being 3 – 5 cm in diameter (WSNWCB 2016). The only means of propagation is by seed (Andersen and Louda 2006), it cannot spread through vegetative means (Andersen and Louda 2006). Plants begin from the seed and establish in the fall of the first year, the plant overwinters in a cluster of radiating leaves at the plant base or as a rosette (Andersen and Louda 2006). It resumes growth in the spring and increases its growth rate during the summer months, producing a large flowering stem (WSNWCB 2016). Flowers are produced from

July to October (Andersen and Louda 2006), and they are cross-pollinated by insects. Seed production is very high, with each thistle producing thousands of seeds (WSNWCB 2016). The seeds are released and dispersed by the wind, and each seed has the capability of been carried long distances by the wind

(WSNWCB 2016).

There are several ways to successfully control C. vulgare. Mechanical control can be successful with several techniques. To prevent seed spread, one can hand-pull individual flowering plants and trash them (WSNWCB 2016).

Mowing can be somewhat effective, if mowing is conducted in concert with flowering. If cut before flowering, the plant may re-sprout and flower again within the same season (WSNWCB 2016). Biological controls are currently in use within the United States and Canada. The seed head fly (Urophora stylata F.), a native of Europe, has been released in United States (WSNWCB 2016). The

7 Texas Tech University, Neil Estes, May 2018 larval form of the insect attack the achenes of the flower head reducing overall seed production (WSNWCB 2016). In sustained populations of C. vulgare it has been shown to significantly reduce seed production (WSNWCB 2016). A strain of weevil (Rhinocyllus conicus) has also been introduced into British Columbia as a biological control of C. vulgare (WSNWCB 2016). Studies have shown that these two controls together have the potential to destroy more seed then either one alone (WSNWCB 2016).

C. arvense (Canada thistle) was introduced to North America in the early colonial period (WSNWCB 2016). The first control legislation for the thistle was enacted in Vermont in 1795 and then the second in New York in 1831 (WSNWCB

2016). In North America at present, C. arvense can be found from about 37° N to about 59° N latitude (Moore 1975). The thistle does not occur in the southern

United States (WSNWCB 2016). C. arvense does best within a temperature range of 0-32° C and with rainfall amounts between 400 mm and 750 mm per year (WSNWCB 2016). C. arvense can grow on a variety of soil types including clay loam, sandy loam, sandy clay and even sand dunes (WSNWCB 2016). It does not do well on wet soils that have poor aeration and prefers areas with abundant light as it grows poorly in shaded areas producing few flowers

(WSNWCB 2016). C. arvense is probably native to southeastern Europe and the eastern Mediterranean (Moore 1975). It can now be found throughout North

America, Europe, North Africa, Asia and Japan (Moore 1975). C. arvense has also been naturalized in South Africa, New Zealand and Australia (Moore 1975).

Other common names include Californian thistle, creeping thistle and field thistle

(USDA 2016).

8 Texas Tech University, Neil Estes, May 2018

C. arvense is a perennial herb commonly found in disturbed areas such as pastures and croplands as well as areas with fluctuating water levels such as stream banks, sedge meadows and wet prairies. The thistle is rhizomatous and has a deep complex root system that spreads horizontally giving rise to aerial shoots that are slender, green, freely branched and 0.3 – 1.2 m tall (WSNWCB

2016). The plant can survive indefinitely and can spread through the root system because of its highly successful vegetative propagation carried on by creeping horizontal roots (WSNWCB 2016). Small pieces of the root can also grow into new plants. Leaves are alternate, sessile, deeply lobed and margined with stiff yellowish spines (WSNWCB 2016). C. arvense produces a large number of small flower heads, in clusters at the tips of branches (WSNWCB 2016). The thistle blooms from June through October. Flowers are purple and dioecious. Four varieties have been recognized based on variation in leaf characteristics, texture, vestiture, segmentation and spininess (WSNWCB 2016). C. arvense is also the only thistle with both male and female plants (Moore 1975).

C. arvense is difficult to control once it is established. Combinations of treatments may be necessary to see any level of successful reduction (WSNWCB

2016). Tilling is the most effective mechanical control of the thistle. Repeating the tillage at 7 to 28 day intervals for up to 4 years can be effective against large infestations of C. arvense (WSNWCB 2016). Mowing can also be effective in reducing seed production in low-level infestations (WSNWCB 2016). Another effective means of control is the planting of competitive crops such as alfalfa and certain types of forage grasses (WSNWCB 2016). A biological control for C. arvense does exist but the agent alone does not kill the plant or prevent its

9 Texas Tech University, Neil Estes, May 2018 spread, the larvae of the stem gall fly (Urophora cardui) can inhibit general growth and flowering of the thistle but have limited effects on established populations (WSNWCB 2016).

C. nutans (musk thistle) is native to southern Europe and western Asia

(WSNWCB 2016). It was introduced to North America in the early 1900’s accidentally, possibly through contaminated ballast water (USDA 2016).

Currently C. nutans is widespread in both Canada and the United States, thriving in disturbed areas such as areas of heavy livestock use and or grazing (USDA

2016). Known by many common names including chardon penche, nodding plumeless thistle, nodding thistle and plumeless thistle (USDA 2016), it can grow at any elevation between sea level to 2,743 meters (Gassman and Kok 2002). The thistle prefers neutral to acidic soils (Gassman and Kok 2002). While it is most successful invading disturbed areas, C. nutans can also invade healthy plant communities (WSNWCB 2016). Unchecked, the thistle can grow up to 2.4 m tall and create impenetrable thickets and crowd out desirable forage plants

(WSNWCB 2016). Livestock will not graze in areas heavily infested with musk thistle (WSNWCB 2016); however, young plants can be grazed by sheep and or goats. Unmanaged grazing will result in larger infestations due to the unpalatable nature of the mature thistle (Popay and Field 1996). C. nutans is not tolerant of excessive wet or dry soils and does does not thrive in shady conditions

(WSNWCB 2016).

C. nutans is generally a biennial plant though it sometimes may only be a winter annual (WSNWCB 2016). In its first fall of growth the thistle forms a rosette and then overwinters. With the return of spring and the start of its

10 Texas Tech University, Neil Estes, May 2018 second year the thistle resumes growth and completes its cycle to seed. C. nutans produces a deep taproot and does not propagate through vegetative means.

Leaves are thick, spiny and lobed with wavy, white outlined margins and light- green veins running through the middle of the leaf. Basal leaves are larger then leaves farther up the stem. Stems are spiny and then winged except just below the flower head (WSNWCB 2016). C. nutans produces large solitary flower heads at the end of each stem. The flowers are reddish purple and droop at maturity.

Each thistle produces 50 to 100 flower heads and each flower head can produce up to 1000 seeds (WSNWCB 2016). Seeds have no plume or parachute so dispersion is gravity and wind driven.

C. nutans can be effectively controlled in improved pasture situations by establishing a dense well-maintained pasture (WSNWCB 2016). It can also be dug or grubbed out as an effective means of control (WSNWCB 2016).

Cultivation will kill musk thistle seedlings and a seed eating weevil, Rhinocyllus conicus, can be effective in reducing seed output (WSNWCB 2016).

The three non-native thistles in the VCNP may have colonized through three mechanisms. Roads may have allowed for the nonnative plant recruitment and spread from neighboring invaded areas (Gelbard and Belnap 2003, Christen and Matlack 2009, Flory and Clay 2009). Thistles often occur along roadsides in the disturbed soil of a drainage ditch or even the road itself and seeds are often located in these same areas. Seeds can be carried along roadways by animals, wind, and over long distances by motorized vehicles (Christen and Matlack

2009). The 1,931 km of roads in the VCNP may have facilitated the distribution of seeds through the preserve and into the areas that are currently being burned

11 Texas Tech University, Neil Estes, May 2018 and or thinned. The restoration processes of burning and thinning can aid in the invasion by nonnative plants (D’Antonio and Meyerson 2002). Only by mapping and monitoring potentially invasive infestations can restoration progress.

Effects of Non-Native Plants on Insect Communities

The invasion of non-native plants into native ecosystems is identified as a major threat to biodiversity and ecosystem structure and function (Vitousek et al.

1997, Mack et al. 2000). Non-native plants can become invasive and affect native plants through competition with and exclusion of native biota. The success of nonnative plants depends on these interactions and those with herbivores

(Maron and Vila 2001). Predicting the consequences of these interactions is one of the greatest challenges of modern land management (Carpenter and Cappccino

2005). Plants form the foundation for most food webs and as a result can affect native insect herbivores, and thus, cascade into larger ecological effects. Non- native plants can affect the quality, abundance, or diversity of native plants thus influencing the abundance or performance of native insects that rely on native vegetation (Elton 1958).

The performance and behavior of native insect herbivores, parasitoids, and predators has evolved over time through their interactions with native biota

(Harvey et al. 2010). Non-native plants can affect the performance and behavior of these insects in many different ways. For example, insect herbivores must be able to discriminate between suitable and unsuitable plants as a means of survival (Chapman 2003). Cues in native plant-insect food webs make it possible for insects to recognize suitable feeding opportunities (Chapman 2003).

12 Texas Tech University, Neil Estes, May 2018

Recognition of suitable oviposition sites is also extremely important and may be linked to plant morphological characteristics (Awmack and Leather 2002).

However, current literature seems to be split on whether or not native insect herbivores perform well on non-native plants. Some studies show non-native plants to be highly suitable hosts for native insects (Harvey et al. 2010, Herrera et al. 2011). On the other hand, other studies show non-native plants to be unsuitable as hosts for native insects, and at times, even toxic to native herbivores (Keeler and Chew 2008, Ding and Blossey 2009, Tallamy et al. 2010).

Exploitation of non-native plants by native herbivores also often differs between specialists and generalists (Chapman 2003). Some studies show that generalists develop well on non-native plants whereas specialists either perished or had very low fitness on the non-native plants (Muller 2009, Harvey et al. 2010a, Fortuna et al. 2012). Predators and parasitoids can also be affected by non-native plants with respect to nutrition of their herbivore prey (Ode 2006). Specifically, nutrients and allelochemicals that are ingested by the herbivore host can affect the feeding and the reproductive ecology of the parasitoid (Ode 2006).

Development of parasitoids is closely linked to their host, and attributes such as size, growth rate, and diet are closely associated with host physiology (Harvey

2005).

It is also possible for native insects to change from native to non-native host plants. Non-native plants may possess unique chemical and morphological traits that are not present in native plant communities and go unrecognized by native herbivores (Harvey et al. 2010). On the other hand, non-native plants can also become an important substitute for native plants. Several studies have

13 Texas Tech University, Neil Estes, May 2018 reported such shifts. For example, a cabbageworm, Pontia protodice, native to the United State feeds on a range of established cruciferous plants that originate in Eurasia (Kingsolver 1985). Other examples can be found in California where

34% of the native butterflies utilized non-native plants as a food source (Graves and Shapiro 2003). Many of these species depend on non-native plants exclusively within urban environments, because the native food plants have been completely eliminated (Shapiro 2002). This ability of a native herbivore to switch from native to a non-native plant depends on the allelochemistry of the invasive plant relative to the native plant. Many plants escape their specialist herbivores when they establish in new habitats. It is only when a new habitat houses closely related species that non-native plants are vulnerable to attack by specialists. For example, native insects that feed on the native thistle, Cirsium altissimum, can readily switch to the non-native C. vulgare because they are closely related (Suwa and Louda 2012). Natural enemies of native herbivores that can make the shift from native to non-native plants, may or may not adopt the non-native substitute, depending upon whether they have the appropriate cues. Predators and parasitoids may colonize non-native plants but this depends on the completion of several hierarchal steps involving the suitability of habitat, plant location, prey/host acceptance, and palatability (Vinson 1976, Isaacs et al.

2009). Little is known about these complex shifts in trophic ecology.

When non-native plants attract native insect herbivores there can also be an evolutionary impact. Insects can quickly utilize these new resources, effectively allowing non-native plants to drive selection resulting in morphological changes (Strauss et al. 2006, Carrol and Fox 2007). If the

14 Texas Tech University, Neil Estes, May 2018 herbivores that are attracted to the non-native plants accept them as a food supply, but have lower fitness than when they utilize the native vegetation, the non-native plants may function as an ecological or evolutionary trap (Keeler and

Chew 2008). These traps can drive selection against the use of non-native plants

(Forister and Scholl 2012). On the other hand, if a non-native plant is more beneficial to native herbivore species then the native insect species may adopt the non-native plant as a host and a shift occurs (Carroll and Fox 2007), leading to morphological and/or physiological adaptations of the insects favoring the non- native plant(s). There have been several studies that show an increase in herbivory over time with respect to non-native plants (Siemann et al. 2006,

Hawkes 2007).

Another interesting way that invasive plants can affect native insects occurs when non-native plants are in close proximity to native plants creating associational effects. If non-native plants attract insect herbivores, attacks on native plants in close proximity may also increase (Russell et al. 2007). The opposite can also be true if non-native plants repel attacks by native herbivores, native plants in close proximity may have reduced occurrences of attack (Haynes and Cronin 2003). Non-native plants also compete with native plants for nutritional resources and light, which can affect the growth and chemistry of the native plant, and can indirectly alter the performance of insects that are associated with the native plants (Russell et al. 2007) by reducing the quality of their host. Plants can also dramatically affect the soil environment they live in.

Plant roots can release a large number of chemicals into the soil that can affect the growth and chemistry of other plants (Bais et al. 2006). If a non-native plant

15 Texas Tech University, Neil Estes, May 2018 releases chemicals into the soil that damages a host plant, the insect-plant interaction between native plants and native herbivores may also be affected

(Bais et al. 2003). There have also been a number of studies showing how non- native plants interact with soil-born pathogens and symbionts (Mangla et al.

2008). Through the effects on soil pathogens and symbionts, non-native plants can affect current and future generations of native plants because both of these biota can induce plant defense mechanisms, which in turn can affect the insect- plant interactions between native insects and native plants (Bezemer and van

Dam 2005).

When a nonnative plant invades a native range, native insect communities can be affected when the non-native plant competes with the native vegetation or displaces the native host plants (Wagner and Van Driesche 2010). These effects are apparent in a loss of native insect diversity or abundance. Studies have shown that a reduction or complete removal of non-native plants leads to an increase or even a full recovery of the native insect community (Gratton and

Denno 2006, Hanula and Horn 2011). Another approach to insect community work and non-native plants has focused on comparing the insect damage on non- native plants and related native plants that occur in the same areas in an attempt to examine the enemy release hypothesis (Keane and Crawley 2002). This hypothesis states that the success of non-native plants in establishment and spread can be attributed to the release from or lack of specialized herbivores and other natural enemies (Keane and Crawley 2002). In their native range, non- native plants are kept in check by their natural herbivores and enemies. In an invaded range, these natural enemies are not present. However, non-native

16 Texas Tech University, Neil Estes, May 2018 plants can experience less, more, or relatively equal amounts of herbivore damage as native plants in the same area depending on the plants in question and the native community within which the native and non-native plants occur.

Pollinators can also be affected by the invasion of nonnative plants. This is of huge importance as many pollinators are keystone species in their respective ecosystems (Kearns et al. 1998). Non-native plants can affect the interaction between native plants and their pollinators as well as having a direct affect on the native communities of pollinators. These affects can be negative because of competition between the native and non-native plants with respect to pollination,

(Lopezaraiza-Mikel et al. 2007) or the effects can be positive because of larger floral assemblages being created by the non-native and native plants (Johnson et al. 2003). Two recent meta-analyses concluded that there was a negative average effect on native flowering plants in the presence of non-native plants as a result of competition for pollinators (Morales and Traveset 2009, Montero-Castaño and

Vila 2012). Non-native plants may actually be more attractive to pollinators then native plants, luring the pollinators away (Morales and Traveset 2009). Non- native plants can produce flowers that are more colorful, larger, or even more flowers per plant (Pysek et al. 2011). The local density of non-native plants can also have an impact on native plant-pollinator interactions (Kaiser-Bunbury and

Muller 2009).

Despite these recent studies, relatively few studies have addressed the direct effects of non-native plants on native pollinator communities (Kearns et al.

1998). There are also several studies that show non-native plants being readily visited by native generalist pollinators suggesting that the effects of non-native

17 Texas Tech University, Neil Estes, May 2018 plants may be less then I thought (Lopezaraiza-Mikel et al. 2007). The amount of overlap between non-native plants and native plants with respect to pollinator visitation depends heavily on the similarity in floral traits between the non-native and native plants (Gibson et al. 2012). The diversity and abundance of pollinators is higher in uninvaded areas or increases after the non-native plants are removed from an area (Hanula and Horn 2011, Thijs et al. 2012).

18 Texas Tech University, Neil Estes, May 2018

Literature Cited

Agee JK, 1993. Fire ecology of Pacific Northwest forests. Washington DC. Island Press.

Allen CD, 1994. Ecological Perspective: Linking Ecology, GIS, and Remote Sensing to Ecosystem Management. Remote sensing and GIS in ecosystem management p 111.

Allen CD, Bresheats DD, 1998. Drought-induced shift of a forest/woodland ecotone: rapid landscape response to climate variation. Proceedings of the National Academy of Sciences 95: 14839-14842.

Allen CD, 2002. Lots of lightning and plenty of people: An ecological history of fire in the upland southwest. In: T.R. Vale (ed.). Fire, Native Peoples, and the Natural Landscape. Washington: Island Press. p143-193.

Allen CD, Anderson SR, Jass RB, Toney JL, Baisan CH, 2008. Paired charcoal and tree-ring records of high-frequency Holocene fire from two New Mexico bog sites. International Journal of Wildland Fire 17: 115–130.

Andersen CP, Louda SM, 2006. Abundance of and floral herbivory on exotic bull thistle versus native tall thistle in western tallgrass prairie. Faculty Publications in the Biological Sciences 74. http://digitalcommons.unl.edu/bioscifacpub/74

Anschuetz KF, Merlan T, 2007. More than a scenic mountain landscape: Valles Caldera National Preserve land use history. Gen. Tech. Rep. RMRS-GTR- 196. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. p277.

Awmack CS, Leather SR, 2002. Host plant quality and fecundity in herbivorous insects. Annual Review of Entomology 47: 817–44.

Bais HP, Vepachedu R, Gilroy S, Callaway RM, Vivanco JM, 2003. Allelopathy and exotic plant invasion: from molecules and genes to species interactions. Science 301: 1377-80.

Bais HP, Weir TL, Perry LG, Gilroy S, Vivanco JM, 2006. The role of root exudates in rhizosphere interactions with plants and other organisms. Annual Review Plant Biology 57: 233-66.

Bezemer TM, van Dam NM, 2005. Linking aboveground and belowground interactions via induced plant defenses. Trends in Ecological and Evolution 20: 617-24.

19 Texas Tech University, Neil Estes, May 2018

Carroll SP, Fox CW, 2007. Dissecting the evolutionary impacts of plant invasions: bugs and beetles as native guides. Global Change Biology 13: 1644-57.

Carpenter D, Cappuccino N, 2005. Herbivory, time since introduction and the invasiveness of exotic plants. Journal of Ecology 93: 315-21.

Chapman RF, 2003. Contact chemoreception in feeding by phytophagous insects. Annual Review of Entomology 48: 455-84.

Christen DC, Matlack GR, 2009. The habitat and conduit functions of roads in the spread of three invasive plant species. Biological Invasions 11: 453- 465.

Covington WW, Moore MM, 1994. Southwestern ponderosa pine forest structure: changes since Euro-American settlement. Journal of Forestry 92: 39-47.

Dahm CW, Geils BW, 1997. An assessment of forest ecosystem health in the Southwest. USDA Forest Service General Technical Report RM-GTR- 295.

D’Antonio C, Meyerson LA, 2002. Exotic plant species as problems and solutions in ecological restoration: a synthesis. Restoration Ecology 10: 703-713.

Ding JQ, Blossey B, 2009. Differences in preference and performance of the water lily leaf beetle, Galerucella nymphaeae, populations on native and introduced aquatic plants. Environmental Entomology 38: 1653-60.

Elton CS, 1958. The Ecology of Invasions by Animals and Plants. London. Methuen

Flory SL, Clay K, 2009. Effects of roads and forest successional age on experimental plant invasions. Biological Conservation 142: 2531-2537.

Forister ML, Scholl CF, 2012. Use of an exotic host plant affects mate choice in an insect herbivore. The American Naturalist 179: 805-10.

Fortuna TM, Vet LEM, Harvey JA, 2012. Effects of an invasive plant on the performance of two parasitoids with different host exploitation strategies. Biological Control 62: 213-20.

Gassman, A, Kok LT, 2002. Musk thistle (nodding thistle). Biological control of weeds in Australia p128.

Gelbard, JL, Belnap J, 2003. Roads as conduits for exotic plant invasions in a semiarid landscape. Conservation Biology 17: 420-432.

20 Texas Tech University, Neil Estes, May 2018

Gibson MR, Richardson DM, Pauw A, 2012. Can floral traits predict an invasive plant’s impact on native plant-pollinator communities? Journal of Ecology 100: 1216-23.

Gratton C, Denno RF, 2006. Arthropod food web restoration following removal of an invasive wetland plant. Ecological Applications 16: 622-31.

Graves SD, Shapiro AM, 2003. Exotics as host plants of the California butterfly fauna. Biological Conservation 110: 413-33.

Hanula JL, Horn S, 2011. Removing an invasive shrub (Chinese privet) increases native bee diversity and abundance in riparian forests of the southeastern United States. Insect Conservation and Diversity 4: 275-83.

Harvey JA, 2005. Factors affecting the evolution of development strategies in parasitoid wasps: the importance of functional constraints and incorporating complexity. Entomologia Experimentalis et applicata 117: 1- 13.

Harvey JA, Biere A, Fortuna T, Vet LEM, Engelkes T, Morren E, Gols R, Verhoeven K, Vogel H, Macel M, Heidel-Fischer HM, Schramm K, van der Putten WH, 2010. Ecological fits, mis-fits and lotteries involving insect herbivores on the invasive plant, Bunias orientalis. Biological Invasions 12: 3045-3059.

Harvey JA, Bukovinszky T, van der Putten WH, 2010. Interactions between invasive plants and insect herbivores: A plea for multitrophic perspective. Biological Conservation 143 10: 2251-2259.

Hawkes CV, 2007. Are invaders moving targets? The generality and persistence of advantages in size, reproduction, and enemy release in invasive plant species with time since introduction. The American Naturalist 170: 832- 843.

Haynes KJ, Cronin JT, 2003. Matrix composition affects the spatial ecology of a prairie planthopper. Ecology 84: 2856-2866.

Heiken G, Goff F, Gardner JN, Hulen JB, Nielson DL, Vaniman, D, 1990. The Valles/Toledo Caldera Complex, Jemez Volcanic Field, New Mexico. Annual Review of Earth and Planet Science 18: 27-53.

Herrera AM, Carruthers RI, Mills NJ, 2011. No evidence for increased performance of a specialist psyllid on invasive French broom. Acta Oecologica 37: 79-86.

Isaacs R, Tuella J, Fiedler A, Gardiner M, Landis D, 2009. Maximizing arthropod-mediated ecosystem services in agricultural landscapes: the

21 Texas Tech University, Neil Estes, May 2018

role of native plants. Frontiers in Ecology and the Environment 7: 196- 203.

Johnson SD, Peter CI, Nilsson LA, Agren J, 2003. Pollination success in a deceptive orchid is enhanced by co-occurring rewarding magnet plants. Ecology 84: 2919-2927.

Kaiser-Bunbury C, Muller C, 2009. Indirect interactions between invasive and native plants via pollinators. Naturwissenschaften 96: 339-346.

Kaufmann MR, Huckaby LS, Regan CM, Popp J, 1998. Forest reference conditions for ecosystem management in the Sacramento Mountains, New Mexico. USDA Forest Service General Technical Report RMRS-GTR-19.

Keane RM, Crawley MJ, 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology and Evolution 17: 164-70.

Kearns CA, Inouye DW, Waser NM, 1998. Endangered mutualisms: the conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 29: 83-112.

Keeler MS, Chew FS, 2008. Escaping an evolutionary trap: preference and performance of a native insect on an exotic invasive host. Oecologia 156: 559-568.

Kingsolver JG, 1985. Thermoregulatory significance of wing melanization in Pieris butterflies (Lepidoptera: Pieridae): physics, posture, and pattern. Oecologia 66: 546-553.

Lopezaraiza-Mikel ME, Hayes RB, Whalley MR, Memmott J, 2007. The impact of an alien plant on a native plant-pollinator network: an experimental approach. Ecology Letters. 10: 539-550.

Mack RN, Simberloff D, Lonsdale WM, Evans H, Clout M, Bazzaz FA, 2000. Biotic invasions: causes, epidemiology, global consequences, and control. Ecological Applications 10: 689-710.

Mangla S, Inderjit, Callaway RM. 2008. Exotic invasive plant accumulates native soil pathogens which inhibit native plants. Journal of Ecology 96: 58-67.

Maron JL, Vila M, 2001. When do herbivores affect plant invasion? Evidence for the natural enemies ` and biotic resistance hypotheses. Oikos 95: 361-373.

Martin C, 2003. Valle Grande: History of the Baca Location Number 1. All Seasons Publishing, Los Alamos, NM, US.

22 Texas Tech University, Neil Estes, May 2018

Montero-Castano A, Vila M, 2012. Impact of landscape alteration and invasions on pollinators: a meta-analysis. Journal of Ecology 100: 884-93.

Morales CL, Traveset A, 2009. A meta-analysis of impacts of alien versus native plants on pollinator visitation and reproductive success of co-flowering native plants. Ecology Letters 12: 716-728.

Moore RJ, 1975. The biology of Canadian weeds.: 13. Cirsium arvense (L.). Canadian Journal of Plant Science 55: 1033-1048.

Muldavin EH, Tonne P, 2003. A vegetation survey and preliminary ecological assessment of Valles Caldera National Preserve, New Mexico, Los Alamos, NM, Valles Caldera Trust, Final Report, Cooperative Agreement No. 01CRAG0014. Available online from the New Mexico Natural Heritage Program at http://nmnhp.unm.edu/vlibrary/pubs_ns/nmnhp_pubs_ns.php.

Muller C, 2009. Role of glucosinolates in plant invasiveness. Phytochemical Review 8: 227-242.

Ode PJ, 2006. Plant chemistry and natural enemy fitness: effects on herbivore and natural enemy interactions. Annual Review of Entomology 51: 163- 185.

Popay I, Field R, 1996. Grazing animals as weed control agents. Weed Technology 10:217-231.

Pysek P, Jarosık V, Chytry M, Danihelka J, Kuhn I, Pergl J, Tichy L, Biesmeijer JC, Ellis WN, Kunin WE, Settele J, 2011. Successful invaders co-opt pollinators of native flora and accumulate insect pollinators with increasing residence time. Ecological Monographs 81: 277-293.

Russell FL, Louda SM, Rand TA, Kachman SD, 2007. Variation in herbivore- mediated indirect effects of an invasive plant on a native plant. Ecology 88: 413-423.

Shapiro AM, 2002. The Californian urban butterfly fauna is dependent on alien plants. Diversity and Distributions 8: 31-40.

Smith JK, 2000. Wildland fire in ecosystems: Effects of fire on fauna. Gen. Tech. Rep. RMRS-GTR-42-vol. 1. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. p83.

Siemann E, Rogers WE, Dewalt SJ, 2006. Rapid adaptation of insect herbivores to an invasive plant. Proceedings of the Royal Society of B 273: 2763-69.

23 Texas Tech University, Neil Estes, May 2018

Strauss SY, Lau JA, Carroll SP, 2006. Evolutionary responses of natives to introduced species: What do introductions tell us about natural communities? Ecology Letters 9: 354-371.

Suwa T, Louda SM, 2012. Combined effects of plant competition and insect herbivory hinder invasiveness of an introduced thistle. Oecologia 169: 467- 476.

Swetnam TW, Baisan CH, 1996. Historical fire regime patterns in the southwestern United States since AD 1700. In ‘Fire Effects in Southwestern Forests: Proceedings of the 2nd La Mesa Fire Symposium’. USDA Forest Service General Technical Report RMGTR-286 p11-32.

Swetnam TW, Betancourt JL, 1998. Mesoscale disturbance and ecological response to decadal climatic variability in the American Southwest. Journal of Climate 11: 3128-3147.

Tallamy DW, Ballard M, D’Amico V, 2010. Can alien plants support generalist insect herbivores? Biological Invasions 12: 2285-2292.

Thijs KW, Brys R, Verboven HAF, Hermy M, 2012. The influence of an invasive plant species on the pollination success and reproductive output of three riparian plant species. Biological Invasions 14: 355-365.

Touchan R, Allen CD, Swetnam TW, 1996. Fire history and climatic patterns in ponderosa pine and mixed-conifer forests of the Jemez Mountains, northern New Mexico. United States Department of Agriculture Forest Service General Technical Report, p33-46.

United States Department of Agriculture (USDA) 2016 Natural Resource Conservation Service (NRCS) http://plants.usda.gov/core/profile?symbol=ciar4

Vinson SB, 1976. Host selection by insect parasitoids. Annual Review of Entomology 21: 109-133.

Vitousek PM, Dantonio CM, Loope LL, Rejmanek M, Westbrooks R, 1997. Introduced species: a significant component of human-caused global change. N. Z. Journal of Ecology 21: 1-16.

Wagner DL, Van Driesche RG, 2010. Threats posed to rare or endangered insects by invasions of nonnative species. Annual Review of Entomology 55: 547- 568.

Washington State Noxious Weed Control Board (WSNWCB) 2016. Written Findings. http://www.nwcb.wa.gov/siteFiles/Carduus_nutans.pdf

Weaver H, 1951. Fire as an Ecological Factor in the Southwestern Ponderosa Pine

24 Texas Tech University, Neil Estes, May 2018

Forests. Journal of Forestry 49: 93-98.

25 Texas Tech University, Neil Estes, May 2018

CHAPTER 2

THISTLE DISTRIBUTION AND OCCURRENCE IN THE VALLES CALDERA NATIONAL PRESERVE, NEW MEXICO, USA

Abstract

Non-native species are an important threat to rangeland health and productivity.

In New Mexico highland range, three non-native plant species of interest,

Cirsium vulgare, Cirsium arvense, and Carduus nutans, are distributed across a patchy and disturbed landscape in 2016 and 2017. I measured non-native species abundance and distribution using road surveys and targeted search efforts. I then correlated non-native species distribution to burn severity data from the

2011 Las Conchas Wildfire and 2013 Thompson Ridge Wildfire, thinning treatments, and aspect at the Valles Caldera National Preserve. Focal species occurred more frequently at burned sites then at unburned sites with 98 clusters of C. vulgare found at burned sites and no clusters found at unburned sites, 13 clusters of C. arvense found at burned sites and no clusters found at unburned sites, six clusters of C. nutans found at burned sites and one cluster found at unburned. We also found eight clusters of C. vulgare and two clusters of C. arvense found in areas undergoing thinning treatments. These results will inform land management plans for non-native species management at disturbed sites such as those impacted by restoration efforts or wildland fire. As fire frequency and intensity increases with changing climates in the southwestern

United States, understanding the relationship between fire mitigation techniques such as thinning, unplanned wildland fires and biological invasions will become increasingly important.

26 Texas Tech University, Neil Estes, May 2018

Introduction

Invasion of native plant communities by non-native species is occurring around the globe at an alarming rate (Vitousek et al. 1997, Mack et al. 2000).

These invasions may represent a major threat to ecological function in our native communities and the Valles Caldera National Preserve (VCNP; Mack et al. 2000,

Vitousek et al. 1997) in northern New Mexico. Fire suppression coupled with poor land management over the last century have altered the structure, composition, and ecology of most of our rangelands and forests (Keane and

Crawley 2002). Many of the management techniques used to combat past land management activities (e.g., thinning and burning), have the potential to facilitate invasion by non-native plants (D’Antonio and Meyerson 2002) due to the disturbance associated with the application of those techniques. The VCNP, in an effort to restore its forests and grasslands and to protect against wildfire, has thinned and burned its rangelands and forests. Roads are another route by which invasion can occur as they can act as a conduit for the spread of nonnative plants into neighboring habitats (Gelbard and Belnap 2003, Christen and

Matlack 2009, Flory and Clay 2009). As a result of past logging and ranching activities within the VCNP, an extensive road system of about 1900 kilometers exists on the property today. Most of those roads are logging roads and have been closed permanently to vehicular traffic, but 320 kilometers of those roads are in active use today by researchers, caldera staff and visitors.

In recent years, several large wildfires have occurred on or burned into the

VCNP. In 2011 the Las Conchas wildfire spread across the eastern borders of the

VCNP. The wildfire burned 63,131 ha, including about 12,140 ha within the

27 Texas Tech University, Neil Estes, May 2018

VCNP.

The specific objectives of this study were to survey VCNP for non-native thistle species in an attempt to identify potentially problematic species within the park and to map the distribution and occurrence of those species. In this study I identified three non-native thistle species that are present in the VCNP (Cirsium vulgare, C. arvense, and Carduus nutans) and mapped their distribution using multiple methods. Thistle distributions and occurrence were examined in relation to aspect, current thinning treatment, fire treatment, and fire severity data from the Las Conchas wildfire (2011) and the Thompson Ridge wildfire of

(2013). I also examined thistle cluster size as a surrogate for time since establishment of the cluster and related cluster size to fire variables.

Materials and Methods

This project was conducted on the Valles Caldera National Preserve, New

Mexico, USA. The VCNP is located in north central New Mexico within the

Jemez Mountain Range about 128 kilometers north of Albuquerque New Mexico.

The preserve is approximately 35,976 ha and ranges in elevation from 2440 m to

3430 m. The VCNP shares most of its border with the Santa Fe National Forest,

Bandelier National Monument, and Santa Clara Pueblo. The unique topographic features of the VCNP are a product of a resurgent caldera that makes up the entire park. The caldera has erupted numerous times in the past with the most

28 Texas Tech University, Neil Estes, May 2018 recent occurring 50,000 - 60,000 years ago. The caldera is roughly 19 kilometers by 22 kilometers in size and forms the main feature of the Jemez Mountain range

(Heiken et al. 1990). Diverse vegetation communities exist within VCNP; dominant vegetation types include montane grasslands to wetlands, riparian forest, to mixed conifer, spruce, fir and aspen. The most common vegetation community type found in the VCNP is montane grasslands. These grasslands reside predominantly in the valley floors while the mixed forest communities exist on the surrounding mountains (Muldavin and Tonne 2003), aspect and elevation primarily drive changes in forest type at these sites.

I surveyed thistles from 1 June – 2 September 2016 and from 1 June – 31

August 2017. In 2016, I visited 67 previously established vegetation sampling sites within the preserve. These sites were situated throughout VCNP and stratified by vegetation type and are described in detail in Humagian (2016).

Three non-native plant species were identified as species of interest in the VCNP through conversation with VCNP personnel and observations made at these sites during the summer of 2016. These species are Cirsium vulgare (bull thistle),

Cirsium arvense (Canada thistle), and Carduus nutans (musk thistle).

In the summer of 2017, I established 160 points within VCNP (Fig. 2.1), stratified based on north/south aspect and four levels of burn severity (high, low, unchanged and unburned; as calculated by dNBR; MTBS.gov). I also established

20 random points within areas that had been thinned but not burned in the wildfires of interest. These points where not stratified in any way other then being in a thinned area, as thinning only occurs on select sites on VCNP. From each point, four 30 meter transects were established in cardinal directions

29 Texas Tech University, Neil Estes, May 2018

(north, south, east, west). I walked transects and recorded the location of any targeted invasive plant found within visual range from of the transect with a GPS in UTMs. I recorded species and a categorical measurement of cluster size

(small, medium, large, extra-large) based on the approximate diameter in meters of the cluster. Small clusters were defined as those < 1 m. Medium clusters were

1-5 m in diameter. Large clusters were 5-10 m in diameter, and extra-large clusters were those 10+ m in diameter.

To supplement data collected from the point method, I also conducted road surveys from 1 June – 6 June 2016 (early growing season), 1 June – 6 June

2017 (early growing season), and 25 – 30 August 2017 (late growing season).

There are approximately 320 kilometers of accessible main access roads in the

VCNP. Due to many of the roads being impassible due to fallen trees, only 101 km of roads were surveyed. A starting point was established at the start of every road to be surveyed and labeled kilometer one. Transects were established perpendicular to the road in one-kilometer increments starting with kilometer one. The direction of the transect, from the road, right or left, was randomly chosen with a coin flip. I then walked the transect out to a randomly chosen length of either 50 meters or 100 meters which was again decided with a coin flip.

The location of any targeted invasive found within visual range was marked with a GPS in UTMs. Both the early growing season and late growing season surveys were conducted on exactly the same roads and exactly the same transects.

I also recorded haphazard detections of target thistle. Any target invasive plant found in a way other then the previously mentioned methods (e.g., chance encounter during travel to and from transects, random encounter during leisure

30 Texas Tech University, Neil Estes, May 2018 time) was marked with a GPS in UTMs (Fig. 2.2). Notes were taken on site with respect to the aspect and the size of the cluster.

I used a contingency analysis to analyze occurrence (defined as 0 = absent,

1 = present; N = 160; alpha = 0.05) of thistle among fire severities, thinning treatments, and aspects. Data were analyzed in JMP Statistical Package (JMP

2007). To visualize my analysis, I created an ordination plot using nonmetric multidimensional scaling (NMDS) with Bray-Curtis Distances run with 2 dimensions. The ordination plots were created using the vegan package in R.

Non-native plants that I encountered in an unstructured way were mapped using

QGIS (QGIS Development Team 2009) and included in the non-metric multidimensional scaling ordination.

Results

At the 180 points established in 2017, target thistle species were present at five burned sites (N = 2 on south aspects, high burn severity, bull thistle, small cluster size; N = 1 south aspect, unchanged burn severity, bull thistle, extra-large cluster size; N = 1 north aspect, low burn severity, musk thistle, small cluster size;

N = 1 north aspect, high burn severity, Canada thistle, small cluster size). Thistle clusters were not detected at unburned points. Thistle clusters were detected at two of the thinned points. Thistle cluster size did not vary by aspect or fire severity (Tables 2.1, 2.2). Thistle occurrence also did not vary by aspect or fire severity (Tables 2.3, 2.4). Additionally, no thistle were detected during any of the three road survey periods.

31 Texas Tech University, Neil Estes, May 2018

I recorded 117 clusters of haphazardously detected target thistles (Fig. 2.2).

These clusters were primarily of bull thistle (N = 97). Canada thistle (N = 13) and musk thistle (N = 7) were also detected in lower abundances. Cluster sizes ranged from small to extra-large, with small clusters being the most common (N

= 70). Medium clusters (N = 34), large clusters (N = 7), and extra-large clusters

(N = 6) were also encountered.

NMDS visualization of C. vulgare, C. arvense, and C. nutans by fitted treatment groups aspect (north, south) and burn severity (high, low, unchanged) showed C. vulgare aligned with south aspect and high severity burn sites while both C. arvense and C. nutans aligned with north and east aspects and low and unchanged burn severity sites (Fig. 2.3).

Discussion

Non-native plant invasions occur through multiple pathways producing small and isolated populations that can be difficult to detect (Barnett et al. 2007).

I experienced similar difficulties in this study. In my first field season, Summer

2016, I visited 67 points within the VCNP. When I stratified points by vegetation type, my thistle detection was very low; however, I noticed patterns in clusters of bull thistle that I discovered when I was en route to these points. The majority of these clusters appeared to be on south aspects in areas subjected to high severity burns during the Las Conchas and Thompson Ridge wildfires. This became the focus of my study, and I examined patterns of thistle occurrence across fire severities and aspects for the three species of interest--bull thistle, Canada thistle, and musk thistle. For my second field season I generated new 180 points to test

32 Texas Tech University, Neil Estes, May 2018 this hypothesis. Detection again proved difficult with these new points; I located only 5 clusters. En route to these points, I again encountered numerous clusters of targeted thistles. By visualizing data with NMDS, I could incorporate the haphazardously encountered thistle data.

The most abundant non-native thistle encountered was bull thistle, followed by Canada thistle and finally musk thistle. The NMDS analysis showed that bull thistle aligned with high severity burned sites on south aspects. The same analysis for Canada thistle and musk thistle did not show a clear alignment of either fire severity or aspect. The NMDS analysis showed their site characteristic preferences as being different then bull thistle, but low sample size prohibited further description. Given our low detection rates for musk and

Canada thistle, it is likely that they are not significant invaders in VCNP.

This study highlights the need for a more comprehensive approach to understanding the complex interactions of invasion. While anthropogenic and natural disturbance have long been associated with the invasion of non-native plants (Macdonald et al. 1988, Rejmánek 1989, Cowie & Werner, 1993, Mack &

D’Antonio, 1998, Gerlach et al. 2003, Mack et al. 2000), there are also other factors affecting their distribution. Precipitation, soil type, and elevation are the fundamental environmental variables controlling plant distribution (Hutchinson

1957, Pysek et al. 2003, Underwood et al. 2004), and the interaction of these ecological factors with disturbances defines a species’ fundamental niche

(Hutchinson 1957, Pysek et al. 2003). Modeling distribution of non-native species with variables such as precipitation, soil type, and elevation may clarify our observed patterns in bull thistle distribution. Such a model might also help

33 Texas Tech University, Neil Estes, May 2018 clarify why Canada thistle and musk thistle seem to prefer areas different then those preferred by bull thistle.

This study also highlights the need for improved detection methods. At the very least, a significantly increased sample size or high resolution sampling grid would potentially increase my ability to detect plants. Given that my unstructured survey produced 122 clusters, it is clear that clusters are present, but finding the clusters in a structured consistent manner that can be analyzed producing meaningful results is difficult. Adaptive cluster sampling is an alternative to conventional random point sampling methods for estimating rare and clustered populations (Thompson 1990, 1991, Roesch 1993). With this method, one searches the vicinity of the original detection in a pattern that spirals out from that original detection. This method predicts that a found cluster predicts the likelihood of another cluster is close by and by spiraling out from that original, the observer can increase the odds of finding more clusters giving you more detections.

Another valuable way to study non-native plant communities is long-term monitoring of trends in occurrence, diversity, abundance, and fitness (Guo 2017).

It is becoming more common for land managers to examine these temporal trends in non-native plant occurrence to supplement current land management decisions and strategies (McLane et al. 2012, Clark et al. 2013). Resource availability is a likely driver of spatial trends (Davis et al. 2000), which vary a great deal after a severe wildfire disturbance. For this reason, it might be better to approach the problem of non-native plant invasion from a successional perspective (Anderson 2007). In California chaparral community, exotic fraction

34 Texas Tech University, Neil Estes, May 2018

(i.e., the proportion of exotic flora to native flora) increases quickly post disturbance and then declines gradually during the initial stages of succession

(Guo 2017). Thus, it may be reasonable to expect a reduction in non-native plant distribution and abundance as succession occurs post-disturbance. It is also reasonable to expect similar trends with disturbances that are associated with land management activities, such as forest thinning. Long-term monitoring of known non-native plant populations is the only way to confirm these trends and develop a long-term perspective on change with respect to land management.

Acknowledgements

The National Park Service and Cooperative Ecosystem Studies Unit and the

Department of Natural Resources Management funded this project. Special thanks to Valles Caldera National Preserve staff, Dr. Bob Parmenter and Ms

Martina Suazo. Thanks to Ms Sharon Smythe for assistance with field work and to Dr. Britt Smith and Mr. Jonathan Knudsen for help in laboratory preparations and analyses.

35 Texas Tech University, Neil Estes, May 2018

Literature Cited

Anderson KJ, 2007. Temporal patterns in rates of community change during succession. The American Naturalist 169: 780-793.

Barnett DT, Stohlgren TJ, Jarnevich CS, Chong GW, Ericson JA, Davern TR, Simonson SE, 2007. The Art and Science of Weed Mapping. Published online: 6 February 2007 # Springer Science + Business Media B.V. 2007.

Christen DC, Matlack GR, 2009. The habitat and conduit functions of roads in the spread of three invasive plant species. Biological Invasions 11: 453- 465.

Clark GF, Johnston EL, Leung B, 2013. Intrinsic time dependence in the diversity invasibility relationship. Ecology 94: 25-31.

D’Antonio C, Meyerson LA, 2002. Exotic plant species as problems and solutions in ecological restoration: a synthesis. Restoration Ecology 10: 703-713.

Davis MA, Grime JP, Thompson K, 2000. Fluctuating resources in plant communities: a general theory of invasibility. Journal of Ecology 88: 528- 534.

Flory SL, Clay K, 2009. Effects of roads and forest successional age on experimental plant invasions. Biological Conservation 142: 2531-2537.

Gelbard JL, Belnap J, 2003. Roads as conduits for exotic plant invasions in a semiarid landscape. Conservation Biology 17: 420-432.

Guo Q, 2017. Temporal changes in native-exotic richness correlations during early post-fire succession. USDA Forest Service Southern Research Station, 3041 Cornwallis Rd., Research Triangle Park, NC 27709, USA.

Heiken G, Goff F, Gardner JN, Hulen JB, Nielson DL, Vaniman D, 1990. The Valles/Toledo Caldera Complex, Jemez Volcanic Field, New Mexico. Annual Review of Earth Planet Science 18: 27-53.

JMP®, Version 12.1.0. SAS Institute Inc., Cary, NC, 1989-2007.

Keane RM, Crawley MJ, 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology and Evolution 17: 164-170.

Mack RN, Simberloff D, Lonsdale WM, Evans H, Clout M, Bazzaz FA, 2000. Biotic invasions: causes, epidemiology, global consequences, and control. Ecological Applications 10: 689-710.

36 Texas Tech University, Neil Estes, May 2018

McLane CR, Battaglia LL, Gibson DJ, Groninger JW, 2012. Succession of exotic and native species assemblages within restored floodplain forests: a test of the parallel dynamics hypothesis. Restoration Ecology 20: 202-210.

Roesch FA Jr, 1993. Adaptive cluster sampling for forest inventories. Forest Science 39: 655-669.

Thompson SK, 1990. Adaptive cluster sampling. Journal of the American Statistical Association 85: 1050-1059.

Thompson SK, 1991. Stratified adaptive cluster sampling. Biometrika 78: 389- 397.

QGIS Development Team, 2009. QGIS Geographic Information System. Open Source Geospatial Foundation. http://qgis.osgeo.org.

Vitousek PM, Dantonio CM, Loope LL, Rejmanek M, Westbrooks R, 1997. Introduced species: a significant component of human-caused global change. New Zealand Journal of Ecology 21: 1-16.

37 Texas Tech University, Neil Estes, May 2018

Table 2.1. Contingency analysis of thistle cluster size (small, medium, large, extra large) by aspect (north, south, east, west). P < 0.05 was considered significant.

Test ChiSquare P Likelihood Ratio 1.356 0.5077 Pearson 0.970 0.6158

Table 2.2. Contingency analysis of thistle cluster size by fire severity. P < 0.05 was considered significant.

Test ChiSquare P Likelihood Ratio 11.746 0.1629 Pearson 10.948 0.2047

Table 2.3. Contingency analysis of thistle occurrence by aspect. P < 0.05 was considered significant.

Test ChiSquare P Likelihood Ratio 0.171 0.6795 Pearson 0.169 0.6806

Table 2.4. Contingency analysis of thistle occurrence by fire severity. P < 0.05 was considered significant.

Test ChiSquare P Likelihood Ratio 6.005 0.1988 Pearson 5.162 0.2710

38 Texas Tech University, Neil Estes, May 2018

POINT MAP | Valles Caldera National Preserve

449 453 S A N T A F E N F 99 27 144 315

144

144 144 D 11

Cerro de Cerro 0901 la Garita Toledo Rito de los Indios 13 144 S A N T A C L A R A V A L 12 L E S A I N D I A N 09 N A N T O N I O R E S E R V A T I O N 10 [K 09 14 08 09 1401

San Anto V nio Creek D 13 A 10 A L L S E 144 O T R O D L E A D T O C F S E 1402 Cerro N a n 14 R N A R A 08 San Luis S n O C t Cerro D o S D n E io Seco S D D R 09 Cr O e E R e R k O E L San Antonio R D O E S F Mountain S C

D D 1403 P 02 O E S A N VALLE 06 L O S O SEC O B

A Y R D

N I

T A G 09 179 O N C 09 05

S A R Sulphur Ck.

U A

S H

P N L A L A M U O [N T S 08 C A N S Y O A N 03 L O I A M I L E J A R R V A L L E D R R Cerro del A E

D Medio F

R E O E

B 03 02 L 0301 O

O D

S D Jaramillo Ck. e s

0201 N d o D O n ad V D D o d 105 A 0302 c

E Redondito n R L Ri L

[K E 03 los Sol

02 N 04 0201 Cerro A D Redondo Grande 02 T

Peak V A L L E G R A N D E I Redondo Ck.

La Jara Ck. O N Redondo

[A A Meadows er 4 01 Riv CD [P z L 02 me

e J ork F ast 02 E

CD4 F O 01

[P [K [A B A N C O B O N I T O 02 R E 0701 %[ El 02 D South S 07 Rabbit Cajete Mountain T Mountain % D 07 [ 289

East 268 Fork [A Jemez B A N D E L I E R Rive 36 r N M S A N T A F E N F 268 289

[ Visitor Contact Station Remote Access Point Points in Burned Areas [ Horse Trailer Parking Hiking/Biking/Equestrian Trail ! Vehicle Entrance Backcountry Vehicle Route Points in Unburned Areas! Restroom Other Roads Primitive Campground Restricted Area Points in Thinned Areas [ Developed Campground ! Parking

Figure 2.1: Location and type of generated points on the VCNP for purpose of thistle sampling.

39 Texas Tech University, Neil Estes, May 2018

FOUND CLUSTER MAP | Valles Caldera National Preserve

449 453 S A N T A F E N F 99 27 144 315

144

144 144 D 11

Cerro de Cerro 0901 la Garita Toledo Rito de los Indios 13 144 S A N T A C L A R A V A L 12 L E S A I N D I A N 09 N A N T O N I O R E S E R V A T I O N 10 [K 09 14 08 09 1401

D D 1403 P 02 O E S A N VALLE 06 L O S O SEC O B

A Y R D

N I

T A G 09 179 O N C 09 05

S A R Sulphur Ck.

U A

S H

P N L A L A M U O [N T S 08 C A N S Y O A N 03 L O I A M I L E J A R R V A L L E D R R Cerro del A E

D Medio F

R E O E

B 03 02 L 0301 O

O D

S D Jaramillo Ck. e s

0201 N d o D O n ad V D D o d 105 A 0302 c

E Redondito n R L Ri L

[K E 03 los Sol

02 N 04 0201 Cerro A D Redondo Grande 02 T

Peak V A L L E G R A N D E I Redondo Ck.

La Jara Ck. O N Redondo

[A A Meadows er 4 01 Riv CD [P z L 02 me

e J ork F ast 02 E

CD4 F O 01

[P [K [A B A N C O B O N I T O 02 R E 0701 %[ El 02 D South S 07 Rabbit Cajete Mountain T Mountain % D 07 [ 289

East 268 Fork [A Jemez B A N D E L I E R Rive 36 r N M S A N T A F E N F 268 289

[ Visitor Contact Station Remote Access Point Cirsium vulgare Cluster [ Horse Trailer Parking Hiking/Biking/Equestrian Trail Cirsium arvense Cluster ! Vehicle Entrance Backcountry Vehicle Route ! Restroom Other Roads Carduus nutans Cluster Primitive Campground Restricted Area [ Developed Campground Bromus Tectorum Cluster ! Parking

Figure 2.2: Locations of found focal thistle by species within the perimeter of

VCNP.

40 Texas Tech University, Neil Estes, May 2018

Figure 2.3: Non-metric multidimensional scaling ordination of focal thistle species (C. nutans, musk thistle; C. vulgare, bull thistle; and C. arsense, Canada thistle) occurrence with fitted treatment groups of burn severity (high, low, unchanged)and aspect (north, south east, west). Stress = 0.092, Dimensions = 2.

41 Texas Tech University, Neil Estes, May 2018

CHAPTER 3

ARTHROPOD COMMUNITIES ON NATIVE AND NON-NATIVE THISTLE IN THE VALLES CALDERA NATIONAL PRESERVE, NEW MEXICO, USA

Abstract

Arthropod community interactions with native and non-native thistle species are an important component of rangeland health and productivity. In New Mexican highland range, one non-native species of thistle, Cirsium vulgare, and one native species of thistle, Cirsium pallidum, are selected for the purpose of comparing arthropod community distribution. I measured arthropod richness and abundance on these plants by collecting 22 C. vulgare thistle plants and 24

C. pallidum thistle plants and then identifying the families and functional groups that were present on each of the plants. One-way ANOVAs were used to compare functional feeding group abundance, taxonomic richness and relative abundance of arthropods by thistle species. I found that spiders and moths/butterflies were more abundant on native thistle. Beetles were more abundant in the native thistle (p = 0.038). In addition, taxonomic richness differed between thistle species, with native thistle having significantly more families (p = 0.036). When

I analyzed functional groups, I found that predators were more abundant (p =

0.040) on native thistles, and pollinators were also more abundant on native thistles (p = 0.056). Arthropods more frequently occurred on native thistles, suggesting that that non-native thistles may benefit from release from natural enemies such as herbivores and sap-feeders. Conversely, non-native plants may suffer from lack of pollination by native insects. These trends may be important

42 Texas Tech University, Neil Estes, May 2018 variables that influence the success of biological invasions, but further studies are necessary to thoroughly understand the role that insects play in limiting or encouraging the establishment and spread of these species.

Introduction

Native arthropods have evolved complex interactions with native biota.

Non-native plants can affect the performance and behavior of arthropods in many different negative ways including reducing their ability to discriminate between suitable and unsuitable plants or to recognize suitable feeding opportunities (Chapman 2003) and identify suitable oviposition sites (Awmack and Leather 2002). Conversely, non-native plants may have no effect or a positive effect on native arthropods (Harvey et al. 2010, Herrera et al. 2011) by acting as extra nectar, pollen, or food sources. Some native arthropods can host shift from native to non-native plants (Keane and Crawley 2002), but non-native plants may also possess unique chemical and morphological traits that hinder or facilitate detection by native arthropods. These unique traits may go unrecognized by native arthropods, or non-native plants can become an important substitute for native arthropods (Keane and Crawley 2002).

Bull thistle (Cirsium vulgare) is a ruderal species that colonizes bare ground, thus it is often found after intense disturbance events (e.g., wildfires, logging, human development; Forcella and Randall 1994). It has the ability to interfere with native and commercial plant growth (Randall and Rejmanek 1993) and is considered invasive in southwestern United States rangelands (Forcella

43 Texas Tech University, Neil Estes, May 2018 and Randall 1994). Biological control agents have been introduced to control bull thistle with mixed ecological effects (Ward et al. 1974, Louda et al. 1997).

Native insects have the capacity to reduce the growth, survival (Guretzky and Louda 1997) and fitness (Louda et al. 1990, Louda and Potvin 1995) of native thistles, and some studies have documented that native insects will also use invasive thistles (e.g., bull thistle) in their diet enough to potentially limit survival of juvenile plants (Guretzsky and Louda 1997). However, few studies examine differences in insect communities between sympatric native and non-native thistles. One notable exception is a study of thistle floral herbivory in tallgrass prairie in which the authors found no differences in floral herbivory between native (tall thistle, Cirsium altissimum) and non-native bull thistle (Andersen and Louda 2006). In this study I compared native arthropod communities between a native thistle, Cirsium pallidum (pale thistle) and a non-native thistle,

C. vulgare (bull thistle). Specifically, I examined arthropod taxonomic richness, relative abundance, and functional feeding group abundance between these two sympatric species of thistle at the Valles Caldera National Preserve.

Materials and Methods

Thistles (N = 22 C. pallidum, N = 24 C. vulgare) were located opportunistically during surveys (see Chapter 1 for survey details). All thistles were located within the Valles Caldera National Preserve, New Mexico, USA (see

Chapter 1 for site description). In recent years, several large mixed severity wildfires have occurred on or burned into the VCNP, increasing its vulnerability to invasion by C. vulgare. In 2011 the Las Conchas wildfire spread across the

44 Texas Tech University, Neil Estes, May 2018 eastern borders of the VCNP. The wildfire burned 63,131 ha, including about

12,140 ha within the VCNP. In 2013 the Thompson Ridge wildfire burned about

9,712 ha. This wildfire’s perimeter was completely within the VCNP. These wildfires have helped shape the land management techniques currently being utilized on the VCNP.

Thistles were sampled by removing the entire above-ground portion of the plant. A trash bag was used to quickly cover the thistle in an effort to maximally capture flying arthropods. The thistle was then cut off at its base and secured in the trash bag. Three cotton balls with 90% ethyl acetate killing agent were placed in the trash bag. The bag was then tied off and used as a modified killing jar.

After arriving back at camp the individual bags were reopened. The thistles inside of them were shaken vigorously for about 20 seconds and then removed from the bag. The contents of the bag were then emptied into a rectangular plastic container (91cm by 60 cm). A small hole was cut in one of the corners of the plastic container. Under the hole I held a 100 ml Nasco Whirl-Pak plastic specimen bag. Each specimen bag contained 90% ethanol. Delicate flying arthropods were placed in a paper pouch instead of the ethanol. The species of thistle, date, time, and location was noted for each thistle that was processed.

All samples were brought back to the lab for further examination under dissecting microscopes (Leica M80). The contents of each specimen bag were emptied onto a petri dish and seed and debris by-catch was discarded.

Arthropods were identified to taxonomic order, and relative abundance and taxonomic richness were calculated. Relative abundance was also calculated for families of interest (e.g., Coccinellidae—a particularly aggressive predatory

45 Texas Tech University, Neil Estes, May 2018 family; Curculionidae—family that includes potential biological control agents).

Functional feeding groups were assigned per Anderson (1995) and included pollinators, plant-dwellers, predators, and herbivores (plant eaters and sap- suckers).

I used a one-way analysis of variance (ANOVA) to compare differences in taxonomic richness and relative abundance of total collected arthropods by thistle type. I also examined relative abundance of individual taxonomic groups and functional group between thistle species using ANOVA. All analyses were conducted in JMP statistical package (JMP 1989-2007). Identifications were verified by Dr. Robin Verble and Dr. Britt Smith.

Results

I observed differences in arthropod communities between native and non- native thistles. Native thistles had a higher relative abundance and taxonomic richness of arthropods than non-native thistles. Several of these differences were non-significant (Table 3.1); however, I observed higher numbers of spiders (p =

0.05, Fig. 3.1), butterflies and moths (p = 0.05, Fig. 3.2), pollinators (p = 0.056,

Fig. 3.9), plant dwellers (p = 0.09, Fig. 3.8) and total arthropods (p = 0.07, Fig.

3.3) on native thistles. I also observed significantly higher taxonomic richness (p

= 0.004, Fig. 3.4), more beetles (p = 0.04, Fig. 3.5), and more predators (p =

0.04, Fig. 3.7) on native thistles.

46 Texas Tech University, Neil Estes, May 2018

Discussion

This study sought to compare differences in insect community structure on native thistle to a non-native thistle. We found more insects (higher relative abundance) and more insect groups (higher taxa richness) on native pale thistles as compared to bull thistle. While previous studies have found that native insects can reduce the growth, survival and fitness of native thistles (e.g., Louda et al.

1990, Louda and Potvin 1995, Guretzky and Louda 1997), this study suggests that similar reductions in growth, survival and fitness may not be possible in our system. Other studies have observed that native thistle insects will use invasive thistles in their diet (Guretzy and Louda 1997), and we found similar results.

Insects are a vital ecosystem component, and differences in community structure may indicate differences in resource availability or detectability. The performance and behavior of native insect herbivores, parasitoids, and predators has evolved over a long period of time through their interactions with native biota

(Harvey et al. 2010). Non-native plants may benefit from decreased predation, as postulated by the enemy release hypothesis (Williamson 1996; Keane and

Crawley 2002). This hypothesis states that since insects that specialize on non- native plants are absent in the introduced range, and that native insect specialists switching to non-native plants is rare, native insect herbivores affect native plants disproportionately more often then non-native plants (Williamson 1996; Keane and Crawley 2002). Our study offers support to this hypothesis; however, we did observe several insects present on bull thistle, suggesting that the rareness of switching to non-native plants may be overestimated.

47 Texas Tech University, Neil Estes, May 2018

This study could be expanded and improved in several ways. Increasing the sample size of both native and non-native thistle specimens would likely clarify observed trends. Identification of the insects to a higher taxonomic resolution also would also improve these data. Collection methods could also be improved to maximize collection of flying insects, which my study underestimates. Supplemental insect netting around target thistle plants may also remedy this deficiency. In addition, inclusion of environmental variables such as soil type, soil moisture, solar radiation, and time of day may clarify these patterns.

Finally, this study spawns several questions that are worthy of further examination. While I examined broad patterns of community composition, other studies have noted different patterns of resource use between generalist and specialist species, with generalists occupying non-native plants more frequently than specialists (Ehrlich and Raven 1965), and this should be examined at VCNP.

In addition, pale thistle produces a yellow flower whereas bull thistle produces a purple flower (Forcella and Randall 1994); differences in flower color can have dramatic consequences for insect detection, and we did not examine this. Finally, plant structure, nectar quality, nectar quantity, perfumes, chemical herbivory deterrents, mechanical herbivory deterrents, life history strategies, and flowering timing (Mygatt and Medeiros 2009, Forcella and Randall 1994) all may differ and can contribute significantly to the insect community structure differences that I observed.

In conclusion, native insects were relatively more abundant and taxonomically diverse on native thistle; however, non-native thistle was also

48 Texas Tech University, Neil Estes, May 2018 occupied by native insects. These differences are important for biological control of non-native thistle and ecologically relevant to conservation of insect biodiversity.

Acknowledgements

The National Park Service and Cooperative Ecosystem Studies Unit and the

Department of Natural Resources Management funded this project. Special thanks to Valles Caldera National Preserve staff, Dr. Bob Parmenter and Ms

Martina Suazo. Thanks to Ms Sharon Smythe for help with fieldwork. Thanks to

Dr. Britt Smith and Mr. Jonathan Knudsen for help with laboratory preparation and analyses.

49 Texas Tech University, Neil Estes, May 2018

Literature Cited

Andersen AN, 1995. A classification of Australian ant communities, based on functional groups which parallel plant life-forms in relation to stress and disturbance. Journal of Biogeography 22: 15-29.

Andersen, CP, Louda SM, 2006. Abundance of and floral herbivory on exotic bull thistle versus native tall thistle in western tallgrass prairie. Prairie Invaders: Proceedings of the 20th North American Prairie Conference. University of Nebraska at Kearney. JT Springer and EC Springer, editors.

Awmack CS, Leather SR, 2002. Host plant quality and fecundity in herbivorous insects. Annual Review of Entomology 47: 817-844.

Chapman RF. 2003. Contact chemoreception in feeding by phytophagous insects. Annual Review of Entomology 48: 455-484

Ehrlich PR, Raven PH, 1965. Butterflies and plants: a study in coevolution. Evolution 19: 586-608.

Forcella F, Randall JM 1994. Biology of bull thistle, Cirsium vulgare (Savi) Tenore. Reviews of Weed Science 6: 29-50.

Guretzky JA, Louda SM, 1997. Evidence for natural biological control: insects decrease survival and growth of a native thistle. Ecological Applications 7: 1330-1340.

Harvey JA, Biere A, Fortuna T, Vet LEM, Engelkes T, Morrien E, Gols R, Verhoeven K, Vogel H, Macel M, Heidel-Fischer HM, Schramm K, van der Putten WH, 2010. Ecological fits, misfits and lotteries involving insect herbivores on the invasive plant, Bunias orientalis. Biological Invasions 12: 3045-3059.

Herrera AM, Carruthers RI, Mills NJ, 2011. No evidence for increased performance of a specialist psyllid on invasive French broom. Acta Oecologica 37: 79-86.

Keane RM, Crawley MJ, 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology and Evolution 17: 164-170.

Louda SM, Potvin MA, Collinge SK, 1990. Predispersal seed predation, postdispersal seed predation and competition in the recruitment of seedlings of a native thistle in sandhills prairie. American Midland Naturalist 124: 105-113.

Louda SM, Potvin MA, 1995. Effects of inflorescence-feeding insects on the demography and lifetime of a native plant. Ecology 76: 229-245.

50 Texas Tech University, Neil Estes, May 2018

Louda SM, Kendall D, Connor J, Simberloff, 1997. Ecological effects of an insect introduced for the biological control of weeds. Science 277: 1088-1090.

JMP®, Version 12.1.0. SAS Institute Inc., Cary, NC, 1989-2007.

Mygatt J, Medeiros J, 2009. Lab manual to the flora of New Mexico

Randall JM, Rejmanek M, 1993. Interference of bull thistle (Cirsium vulgare) with growth of ponderosa pine (Pinus ponderosa) seedlings in a forest plantation. Canadian Journal of Forest Research 23: 1507-1513.

Ward RH, Pienkowski RL, Kok LT, 1974. Host specificity of the first-instar of Ceuthorhynchidius horridus a weevil for biological control of thistles. Economic Entomology 67: 735-737.

Williamson M, 1996. Biological Invasions. Chapman & Hall. London.

51 Texas Tech University, Neil Estes, May 2018

Table 3.1. One-way ANOVA of arthropods by thistle type. DF = degrees of freedom. SS = sum of squares. MS = mean square. *=significant (p < 0.05). Abundance is the number of individuals per plant (i.e., relative abundance). Arthropod richness is the number of orders present on each individual plant.

Source DF SS MS F P Formicidae Abundance 1 14.42951 14.4295 0.7869 0.3798 1 14.42951 14.4295 0.7869 0.3798 Thistle Type 44 806.78788 18.3361 Error 45 821.21739 C. Total

Araneae Abundance 1 19.03821 19.0382 3.9740 0.0524 1 19.03821 19.0382 3.9740 0.0524 Thistle Type 44 210.78788 4.7906 Error 45 229.82609 C. Total

Aphididae Abundance 1 1961.364 1961.36 1.9926 0.1651 Thistle Type 44 43310.049 984.32 Error 45 45271.413 C. Total

Coccinellidae Abundance 1 0.4802372 0.480237 2.8346 0.099 Thistle Type 3 44 7.4545455 0.169421 Error 45 7.9347826 C. Total

Curculionidae Abundance 1 46.43626 46.4363 2.2621 0.1397 Thistle Type 44 903.21591 20.5276 Error 45 949.65217 C. Total

Other Coleoptera Abundance 1 14.42951 14.4295 2.6422 0.1112 Thistle Type 44 240.28788 5.4611 Error 45 254.71739 C. Total

52 Texas Tech University, Neil Estes, May 2018

Table 3.1, Continued…

Apidae/Bombidae Abundance 1 0.0133399 0.013340 0.1321 0.7180 Thistle Type 44 4.4431818 0.100981 Error 45 4.4565217 C. Total

Hymenoptera-other Abundance 1 1.393939 1.39394 2.1822 0.1467 Thistle Type 44 28.106061 0.63877 Error 45 29.500000 C. Total

Lepidoptera Abundance 1 0.976449 0.976449 3.9206 0.054 Thistle Type 0 44 10.958333 0.249053 Error 45 11.934783 C. Total

Diptera Abundance 1 0.016469 0.016469 0.028 0.866 Thistle Type 7 4 44 25.287879 0.574725 Error 45 25.304348 C. Total

Orthoptera Abundance Thistle Type 1 0.0164690 0.016469 0.2599 0.6127 44 2.7878788 0.063361 Error 45 2.8043478 C. Total

Larvae Abundance 1 13.94615 13.9461 2.6451 0.1110 Thistle Type 44 231.98864 5.2725 Error 45 245.93478 C. Total

53 Texas Tech University, Neil Estes, May 2018

Table 3.1, Continued…

Other Arthropoda 1 0.005929 0.005929 0.0142 0.9057 Thistle Type 44 18.363636 0.417355 Error 45 18.369565 C. Total

Hemiptera Abundance 1 0.10293 0.10293 0.0187 0.8917 Thistle Type 44 241.54924 5.48976 Error 45 241.65217 C. Total

Total Arthropod Abundance 1 4960.169 4960.17 3.4988 0.0681 Thistle Type 44 62378.288 1417.69 Error 45 67338.457 C. Total

Arthropod Richness 1 48.55879 48.5588 9.4373 0.003 Thistle Type 6* 44 226.39773 5.1454 Error 45 274.95652 C. Total

Total Beetle Abundance 1 127.8263 127.826 4.5655 0.038 Thistle Type 2* 44 1231.9129 27.998 Error 45 1359.7391 C. Total

Total Hymenoptera Abundance Thistle Type 1 25.95669 25.9567 1.4551 0.2342 44 784.91288 17.8389 Error 45 810.86957 C. Total

54 Texas Tech University, Neil Estes, May 2018

Table 3.1, Continued…

Total Hemiptera Abundance 1 1989.884 1989.88 1.9566 0.1689 Thistle Type 44 44748.833 1017.02 Error 45 46738.717 C. Total

Plant Dweller Abundance 1 3625.636 3625.64 3.066 0.086 Thistle Type 0 9 44 52031.864 1182.54 Error 45 55657.500 C. Total

Pollinator Abundance 1 5.820817 5.82082 3.8637 0.0557 Thistle Type 44 66.287879 1.50654 Error 45 72.108696 C. Total

Predator Abundance 1 25.56588 25.5659 4.4790 0.040 Thistle Type 0* 44 251.15152 5.7080 Error C. Total 45 276.71739

Herbivore Abundance 1 2657.492 2657.49 2.3889 0.1294 Thistle Type 44 48946.943 1112.43 Error 45 51604.435 C. Total

55 Texas Tech University, Neil Estes, May 2018

Figure 3.1. Araneae relative abundance per plant (Y) marginally differs between native and non-native thistle (p = 0.05). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

Figure 3.2. Lepidoptera relative abundance (Y) is marginally different between native and non-native thistle (p = 0.05). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

56 Texas Tech University, Neil Estes, May 2018

Figure 3.3. Arthropod relative abundance (Y) is marginally different between native and non-native thistle (p = 0.07). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

Figure 3.4. Arthropod richness (i.e., the number of individual orders found on each plant; Y) is marginally different between native and non-native thistle (p = 0.004). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

57 Texas Tech University, Neil Estes, May 2018

Figure 3.5. Coleoptera relative abundance (Y) differs between native and non- native thistle (p = 0.04). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

Figure 3.6. Hymenoptera relative abundance (Y) differs between native and non-native thistle (p = 0.04). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

58 Texas Tech University, Neil Estes, May 2018

Figure 3.7. Predator relative abundance (Y) differs between native and non- native thistle (p = 0.04). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

Figure 3.8. Plant dweller relative abundance (Y) differs between native and non-native thistle (p = 0.09). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

59 Texas Tech University, Neil Estes, May 2018

Figure 3.9. Pollinator relative abundance (Y) differs between native and non- native thistle (p = 0.056). Cirsium pallidum = native thistle. Cirsium vulgate = non-native thistle. Bars represent standard error.

60 Texas Tech University, Neil Estes, May 2018

CHAPTER 4

MANAGEMENT RECOMMENDATIONS

Semi-arid forest fire regimes have been dramatically altered over the past two centuries leading to an increase in fuel loads creating a trend of infrequent, large, high severity wildfires (Covington and Moore 1994, Fule et al. 1997, Hunter et al. 2006). This trend is evident when looking at the recent wildfires and the unburned portions of the VCNP. High severity wildfires kill trees, reduce understory vegetation cover and reduce the recovery potential of the seed bank stored within the soil through mortality of the seeds (Korb et al. 2004, Hunter et al. 2006). Within the VCNP there is a visually detectable difference in recovery between areas that experienced high severity burns and areas that experienced low severity burns; in high severity burn sites, areas with bare soil and little to no tree canopy are still found. In contrast, areas subjected to a low severity burn appear to be much further along in the recovery process (e.g., little to no bare soil, intact tree canopy, intact understory vegetation canopy). The potential for establishment of non-native plants increases with a reduction of tree canopy cover and an increase in exposed bare soil (Hunter et al. 2006).

Non-native plant species are a threat to the ecological function of native plant communities around the globe (Vitousek 1997, Mack 2000), and the native plant communities of the Valles Caldera National Preserve are not immune.

Three non-native species of thistle, Cirsium vulgare, C. arvense, and Carduus nutans, have been identified as potentially problematic by VCNP staff. The

VCNP is approximately 35,000 ha of fairly rugged topography ranging from 2440

61 Texas Tech University, Neil Estes, May 2018 to 3430 m in elevation. It is surrounded by and shares most of its border with the

Santa Fe National Forest. Like most of our rangelands, the VCNP and Santa Fe

National Forest been subjected to fire suppression, overgrazing, and anthropocentric land management practices through most of the twentieth century which have altered the structure, composition, and ecology of the respective landscapes (Keane 2002). These issues coupled with modern management techniques have helped to facilitate invasion by non-native plants

(D` Antonio 2002).

With a rich history of logging and ranching, the VCNP has an extensive system of roads totaling about 1900 kilometers in all. Roads are a known route by which invasion can spread as they can aid in the spread of non-native seed from neighboring habitats by way of wind, increased animal traffic, and increased vehicular traffic (Gelbard 2003, Christen 2009, Flory 2009). I conducted road surveys for non-native thistle at VCNP and found no evidence that this is a current mechanism of local spread.

The VCNP also has an extensive wildland fire history. Two recent wildfires that occurred on the VCNP, Los Conchas in 2011 and Thompson Ridge in 2013 represent a very dramatic disturbance on the landscape due to fire suppression and the subsequent fuel accumulation. These types of disturbances are known to facilitate the invasion of non-native plants (D` Antonio 2002) and for that reason

I surveyed areas that had not burned and areas that had burned. I then correlated non-native species distribution to fire treatment, burn severity, and aspect with none being significant. Non-native plant species occurred more frequently at burned sites then at unburned sites, with high severity

62 Texas Tech University, Neil Estes, May 2018 sites having the highest occurrence of invasive plants, though few plants were found in this survey.

Insect community interactions with native and non-native thistle species are also an important component of rangeland health and productivity. Non- native plants can affect native insects in many negative ways as well as many positive ways (Awmack 2002, Keane 2002, Chapman 2003, Harvey 2010,

Herrera 2011). I was particularly interested in insect communities of native versus non-native thistle. One non-native species of thistle, Cirsium vulgare, and one native species of thistle, C. pallidum, were sampled for the purpose of comparing insect taxonomic richness and abundance between the two thistle species. I collected 22 C. vulgare (bull thistle) thistle plants and 24 C. pallidum

(pale thistle) thistle plants and then sampled and identified the insect families and functional groups present on those plants. I found more insects and more insect families and orders on pale thistle (native) than on bull thistle (non-native).

Eradication of non-native thistle may not be practical, even at small spatial scales (WSNWCB 2016). While individual clusters can be managed post- emergence, preventing the disturbance (i.e., wildland fire) that originally increased the likelihood for invasion may be unavoidable in a rapidly changing climate. I recommend that the VCNP monitor non-native clusters to gather data on temporal trends in their size, distribution, and abundance, to further our understanding of long-term trends in plant success post-wildfire. This knowledge is vital when considering future land management.

63 Texas Tech University, Neil Estes, May 2018

Literature Cited

Awmack CS, Leather SR, 2002. Host plant quality and fecundity in herbivorous insects. Annual Review of Entomology 47: 817–44.

Chapman RF, 2003. Contact chemoreception in feeding by phytophagous insects. Annual Review of Entomology 48: 455-84.

Christen DC, Matlack GR, 2009. The habitat and conduit functions of roads in the spread of three invasive plant species. Biological Invasions 11: 453- 465.

Covington WW, Moore MM, 1994. Southwestern ponderosa pine forest structure: changes since Euro-American settlement. Journal of Forestry 92: 39-47.

D’Antonio C, Meyerson LA, 2002. Exotic plant species as problems and solutions in ecological restoration: a synthesis. Restoration Ecology 10: 703-713.

Flory SL, Clay K, 2009. Effects of roads and forest successional age on experimental plant invasions. Biological Conservation 142: 2531-2537.

Fule PZ, Covington WW, Moore MM, 1997. Determining reference conditions for ecosystem management of southwestern ponderosa pine forests. Ecology Applications 7: 895-908.

Gelbard, JL, Belnap J, 2003. Roads as conduits for exotic plant invasions in a semiarid landscape. Conservation Biology 17: 420-432.

Herrera AM, Carruthers RI, Mills NJ, 2011. No evidence for increased performance of a specialist psyllid on invasive French broom. Acta Oecologica 37: 79-86.

Hunter ME, Omi PN, Martinson EJ, Chong GW, 2006. Establishment of non- native plant species after wildfire: effects of fuel treatments, abiotic and biotic factors, and post-fire grass seeding treatment. International Journal of Wildland Fire 15: 271-281.

64 Texas Tech University, Neil Estes, May 2018

Keane RM, Crawley MJ, 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology and Evolution 17: 164-70.

Korb JE, Johnson NC, Covington WW, 2004. Slash pine burning effects on soil biota and chemical properties and plant establishment: recommendations of amelioration. Restoration Ecology 12: 52-62.

Mack RN, Simberloff D, Lonsdale WM, Evans H, Clout M, Bazzaz FA, 2000. Biotic invasions: causes, epidemiology, global consequences, and control. Ecological Applications 10: 689-710.

Vitousek PM, Dantonio CM, Loope LL, Rejmanek M, Westbrooks R, 1997. Introduced species: a significant component of human-caused global change. N. Z. Journal of Ecology 21: 1-16.

Washington State Noxious Weed Control Board (WSNWCB) 2016. Written Findings. http://www.nwcb.wa.gov/siteFiles/Cirsium_arvense.pdf