Black Carpenter Ants in the Ozark Mountains of Arkansas: Relationships with Prescribed Fire, Site and Stand Variables, and Red Oak Borer

Home , Black carpenter ant, Camponotus laevigatus, Carpenter ant, Xenodusa

BLACK CARPENTER ANTS IN THE OZARK MOUNTAINS OF ARKANSAS: RELATIONSHIPS WITH PRESCRIBED FIRE, SITE AND STAND VARIABLES, AND RED OAK BORER

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Entomology

ROBIN MICHELLE VERBLE University of Southern Indiana Bachelor of Science in Biophysics, 2006

August 2008 University of Arkansas

ABSTRACT

Black carpenter ants, Camponotus pennsylvanicus DeGeer, are nearly ubiquitous in eastern North American forests. These ants are documented as predators of red oak borer, Enaphalodes rufulus Haldeman, a native longhorn beetle that underwent an unprecedented population increase and decline in the oak hickory forests of the

Ozark Mountains of Arkansas from the late 1990’s to 2005.

My objective was to examine red oak borer emergence holes and site aspects and correlate these forest and tree attributes with presence or absence of black carpenter ants. Historic red oak borer population data, tree REP class and site aspects for 13 separate plots were used. At each site, all red oaks >10 cm DBH were baited for black carpenters ants using a mixture of tuna in oil and honey. Black carpenter ants are more likely to be found on trees with low levels of previous red oak borer infestation versus those trees with previously high levels of infestation.

These results may suggest black carpenter ants play a role in controlling red oak borer populations. Distribution of black carpenter ants in red oaks prior to and during the outbreak is unknown. Future investigations should be directed at efforts to understand whether black carpenter ants simply prefer different apparently healthy trees or if, via predation, these ants are acting as agents of red oak borer control.

carpenter ants were more commonly present on oaks than on hickories and appear to prefer large trees over small trees. Time elapsed since a prescribed burn appears to be important in determining black carpenter ant presence, potentially via fire- induced habitat modifications, although further investigation is necessary to either confirm or refute this hypothesis.

This thesis is approved for recommendation to the Graduate Council

Thesis Director:

______Dr. Frederick M. Stephen

Thesis Committee:

______Dr. Timothy J. Kring

______Dr. Lynne C. Thompson

THESIS DUPLICATION RELEASE

I hereby authorize the University of Arkansas Libraries to duplicate this thesis when needed for research and/or scholarship.

Agreed______Robin Verble

Refused______Robin Verble

ACKNOWLEDGEMENTS

First and foremost I would like to thank Dr. Fred Stephen. His personal and professional advice, seemingly endless patience, support and guidance as I found my way through the processes of graduate school have been invaluable. I would also like to thank Dr. Tim Kring for his enthusiasm, suggestions and honest advice.

Dr. Lynne Thompson has provided a great deal of insight into my project and I owe him great thanks for his advice and helpful suggestions. Dr. Rob Wiedenman, thank you for your spirited support of my outreach endeavors, professional advice and the many opportunities to expand my horizons.

Most notably, I would like to thank my field assistants, Tyler CarlLee and

Matthew McCall for their willingness to work bizarre hours, remarkable patience and unwavering enthusiasm. Forest entomology laboratory members, John Riggins,

Laurel Haavik and Larry Galligan are owed great thanks for listening, discussing ideas and not complaining (too loudly) about the stinky tuna and honey baits. Mike

Melnechuk, thank you, not only for your continued friendship, but for helping me understand fire effects, locate plots and solidifying my love of prescribed fire.

I would like to thank my father, Mark Verble, for inspiring me to a career in science and believing in my dreams, and my mother, Karen Verble, for her unconditional love and many long hours of telephone conversations. I suppose I should thank my little brother, Todd Verble, for being pretty cool. Megan and Kent

Swedlund, thank you for all the phone calls, visits, and cards. Finally, Seth Pearson, thank you for your unwavering confidence in me, the lengthy discussions about the merits of prescribed fire and for always believing there is a way to get the cat out of the box.

vivi TABLE OF CONTENTS

Page

Thesis Abstract………………………………………………… ii

Acknowledgements…………………………………………… vi

Table of Contents……………………………………………… vii

Chapter 1—Introduction & Literature Review………………. 1

Chapter 2— Occurrence of black carpenter ants in trees previously infested with red oak borer

Introduction……………………………………………… 33 Methods…………………………………………………….. 35 Results…………………………………………………...... 38 Discussion………………………………………………….. 38 Tables & Figures…………………………………………… 41 References………………………………………………… 44

Chapter 3— Occurrence of black carpenter ants in oak-hickory forests in relation to prescribed fire and site-stand variables

Introduction……………………………………………….. 48 Methods…………………………………………………… 51 Results……………………………………………………... 53 Discussion…………………………………………………. 55 Tables & Figures…………………………………………... 59 References…………………………………………………. 62

Chapter 4—Summary and future directions……………. 66

viii

CHAPTER 1

LITERATURE REVIEW

1 1 OZARK MOUNTAINS OF ARKANSAS

The Ozark Highlands are a physiographic province extending from south central

Missouri south and west into Arkansas and Oklahoma at elevations of 75-750 meters above sea level (Read 1952). The Ozarks are characterized by limestone, sandstone and shale rocks, sometimes exposed, and shallow, undeveloped poor soils (Beilmann and

Brenner 1951). Rocks of the Ozark Highlands are sedimentary and form a broad dome that has been highly eroded since the Paleozoic era (Read 1952). Other than alluvial stream bottoms, soils of the region are formed by weathering rock (Read 1952).

Drainage-limiting fragipan soils are present on broad ridgetops (Jenkins and Pallardy

1995).

The Ozark Plateau of Arkansas receives an average of 77-110 centimeters of precipitation per year. Of this, 40% of the precipitation is received in the spring, 33% in the summer and fall, and 27% in the winter (Read 1952, NOAA 2007). On average 199 days per year are frost free. The mean temperature in January is 2.2 C, and the mean

July temperature is 23.9 C (Read 1952, NOAA 2007). The forests of the Ozarks are oak- hickory dominated with significant components of oak-pine and pine. The average age range for trees in Ozark forests is 35-75 years (Loewenstein et al. 2000). Within the

Ozark Highlands, the Ozark National Forest is located in north central and northwestern

Arkansas and comprises 4900 square kilometers (Anonymous 2008).

The modern Ozarks are heavily forested with few natural prairie openings.

Using General Land Office notes, Foti (2004) estimated that the Boston Mountain region of the Ozark Plateau historically had 129 trees over 13.2 cm DBH per hectare. Today, there are 378. trees >13.2 cm DBH per hectare in this region. This near tripling in tree

2 density is attributed to fire suppression in the region throughout much of the twentieth century (Foti 2004). Historic accounts of the region describe a much more open landscape with a greater percentage of unforested land (Palmer 1921). Early explorers report a scarce quantity of timber on the hills and many “balds” and “barrens” (Palmer

1921).

The Ozark Plateau was heavily logged in the late nineteenth century and early twentieth century. Regeneration that occurred after the widespread deforestation resulting from timber harvesting created even-aged stands (Jenkins and Pallardy 1995).

After many areas were logged, severe fires were allowed to burn through the areas with the intention of promoting oak regeneration (VanLear 2004). These fires were halted in the late 1920’s with the enactment of the fire protection and suppression laws (Fralish

2004).

Total suppression of fire in the region has dramatically altered the landscape, allowing mesophytic species such as maple and ash to become more prevalent in an area that was once dominated by oak and hickory. Oak and hickory seedlings rarely reach sapling size in modern forests, due to competition for light (Fralish 2004). In 1999 the

Ozark Highlands experienced a red oak decline event that severely affected (>75% mortality and dieback) 300,000 hectares of the Ozark-St. Francis National Forest

(Heitzman 2003).

OAK DECLINE

Oak decline is a “progressive and general dieback from the tips of branches” caused by a complex interaction of environmental stresses, pests and site factors”

3 (Wargo et al. 1983). Historically, oak decline events have been observed throughout much of the eastern United States, including the Ozark Plateau of Arkansas and

Missouri. In 1996, it was estimated that 1.6 million hectares of oak forests in the southeastern United States were affected by decline (Oak et al.1996).

Manion (1992) developed a forest decline model that includes predisposing factors, inciting factors and contributing factors. Predisposing factors include advanced tree age or poor soil conditions that create a general pervasive stress on trees. Short- term inciting factors include drought and defoliation, which push stressed trees into a

state of decline. Finally, contributing factors including root rot, canker fungi and wood-

infesting insects become evident in the last stages of decline (Heitzman 2003).

Decline events can also be the result of even-aged stands reaching the stage of cohort senescence simultaneously (Muller-Dombois 1992). Cohort senescence is a stage of stand maturity which is coupled with decreased energy as a result of both intrinsic genetic factors and extrinsic stresses. In this situation, decline is not viewed as a disease, but rather as a normal part of the stand ecology (Muller-Dombois 1992).

Members of the red oak group (Quercus: section Lobatae) are more susceptible to decline than white oaks (Heitzman 2003). Red oak decline is specifically characterized by progressive terminal dieback on the branches and bole followed by sudden foliage wilt and browning. Fans and rhizomorphs of Armillaria mellea and galleries and exit

holes of two-lined chestnut borer are present in the final stages of decline (Wargo et al.

1983). Decline usually spreads through a geographic range; peak tree mortality occurs

2-5 years after the initial stress (Wargo et al. 1983). The highest mortality is observed in

older trees and on dry sites (Heitzman 2003).

4 In 1959 and again from 1980-1981, the Ozark Mountains of Arkansas experienced oak mortality events. Another event was recorded in the Missouri Ozarks from 1980-

1986 (Spetich 2004). During the 1999 oak decline event, northern red oak (Quercus rubra

L.), black oak (Q. velutina Lam.) and scarlet oak (Q. coccinea Münchh.) all exhibited widespread dieback and mortality, although the exact causes of mortality are unknown

(Heitzman 2003).

Timber harvesting at the beginning of the twentieth century produced even aged stands of trees that may have been susceptible to cohort senescence and predisposed oaks to widespread decline (Ozark-St Francis National Forest 1978). Using the Manion forest decline model, the 1999 Ozark red oak decline event can be viewed as a result of predisposing factors such as tree age, prolonged drought, poor shallow soils with high clay content, slope and gradient. The event may have been incited by short-term drought stress (Stephen et al. 2001). Finally, red oak borer, Enaphalodes rufulus

(Haldeman) infestation and root rot fungus, Armillaria spp. may have contributed to the final decline of the trees (Stephen et al. 2001). However, because so many potential factors are involved in oak decline, it is difficult to identify the relative of importance of individual factors (Pedersen 1998).

This oak decline event has economic and ecological consequences. Salvage of dead oaks has been a priority, as this decline event has had a substantial economic impact on the timber industry (Starkey et al. 2000). Widespread oak mortality may also have severe ecological consequences for wildlife, because mast production will be reduced (Smith 2006). Restoration goals include the development of silvicultural prescriptions to maintain and improve the long-term health of oak ecosystems, improved red oak regeneration and increased species diversity (Stephen et al. 2001). To

5 minimize both the economic and ecological effects of this red oak decline event, an action plan was created (Starkey et al. 2000). The three steps of this plan are: (1) a significant reduction in overstory basal area, (2) a waiting period of three to five years to

allow existing oak regeneration to develop a more vigorous root system and (3)

prescribed fire to favor the fire-tolerant sprouting oak regeneration over other species

(Heitzman 2003).

PRESCRIBED FIRE

Of the many factors that shaped hardwood forests in eastern North America, fire was perhaps the most important (Batek et al. 1999). Native Americans began inhabiting south-central North America around 12000 years ago and used fire to concentrate game species, promote fruit and berry production, expose nuts, and to open major corridors of travel: Early European settlers continued to use fire for similar reasons—to clear underbrush and to clear agricultural fields. These fires were largely responsible for the creation of an oak-dominated system (VanLear 2004). During the early twentieth century there were serious efforts to exclude fire in eastern North America; however, burning was still common in the Ozark Highlands until the 1940’s (VanLear 2004).

Before the twentieth century, fires in the Ozark Highlands were light to moderate intensity surface fires; however, nearly complete fire exclusion for over 70 years has changed the nature of these fires and allowed for the invasion of oak dominated stands by shade tolerant, fire intolerant species (Dey and Hartman 2005).

Before European settlement of the Ozark Highlands, the fire regime was 4.6-16 years; during the European American period, the fire regime was reduced to 2-3 years. During

6 fire suppression, fire intervals were drastically extended to 62-80 years (Guyette and

Spetich 2003). Fire frequency is positively related to low human population densities; fires occur less frequently at high human population densities (Guyette and Spetich

2003).

The first national USDA Forest Service conference, the Mather Field Conference of 1921 dealt with methods to suppress fire (Saveland 1995). The advent of fire- suppression within forest management coincided with the rise in popularity of the traditional four-stage successional theory in which ecosystems move linearly, from primeval conditions toward a single, undisturbed evolutionary climax ecosystem

(Jurney et al. 2004). Clements (1920) suggested that only undisturbed wilderness areas were suitable representatives for ecological studies. In the traditional linear succession

model, fire is viewed as a disturbance factor holding ecosystems in a sub-climax state

(Jurney et al. 2004).

In 1949, Aldo Leopold suggested a different definition of land health, defining it as “a vigorous state of self renewal” (Leopold 1949); thus triggering debates about the effects of fire. The two opposing schools of thought debated whether fire was destructive and should never be applied by humans or an essential management tool to restore forests to their natural conditions (Jurney et al. 2004). Foresters were initially reluctant to use prescribed fire in hardwood stands, because of fear of damaging boles of high-value trees (Brennan and Hermann 1994). This debate continued until it was finally demonstrated that the linear climax ecosystem model was inferred on theoretical grounds and not based on scientific evidence, bringing prescribed fire back into the scope of ecosystem management (Jurney et al. 2004).

7 With the reemergence of prescribed fire, a strong emphasis has been placed on understanding the dynamic interactions, both biotic and abiotic, that occur prior to and as a result of fire ignition. Prescribed fire is defined by the Bureau of Land Management as “any fire ignited by management actions under certain, predetermined conditions to meet specific objectives related to hazardous fuels or habitat improvement, with a written approved plan” (Anonymous 2007). Multiple factors are at work in the successful ignition and treatment of a given area. The fire environment includes the conditions, influences and modifying forces that control the behavior of fire

(Countryman 1972). The way a fire progresses over the land can be greatly influenced by topography (Countryman 1972). Topography includes aspect, slope, elevation, and configuration of the land in relation to time. The fire environment is also influenced by the vegetation layer and available fuels. This is the source of thermal energy and the driving force behind fire behavior. Characteristics such as fuel loading, porosity, distribution, continuity, chemical composition and moisture content are highly dynamic and are highly variable across a landscape. The final factors influencing the fire environment are elements of weather, such as wind speed, relative humidity, cloud cover, precipitation, and air stability. Weather is the most variable of all the fire environment components (Countryman 1972).

Flammability is the interaction between the fuel, topography and air mass over time (Habeck and Mutch 1973). Flammability has general seasonal trends. Most natural fires tend to occur in spring and summer as a result of lightning strikes (USDA Forest

Service 2000); however, most prescribed fires are ignited during the dormant season when flammability and escape hazard are lower (Clark 1990). Dormant season fires typically create a diverse understory with diverse grasses, legumes and forbs (Brockway

8 and Lewis 1997). Spring and summer growing season fires are particularly effective in killing lower understory trees and reducing shade, thus providing growing space for regenerating oak (Brennan and Hermann 1994).

Root collar diameter and location affect the response of hardwood regeneration to fire (Brose and VanLear 2004). Oak has hypogeal germination in which the cotyledons remain in the root collar below the soil surface insulating them from fire

(Brose et al. 1999). Fire can help prepare the seed bed for acorn caching by wildlife, xerify soil structure to inhibit establishment of mesophytic species, and decrease populations of acorn-infesting insects (Brose 2004). Oaks are three times more abundant in areas that have been treated with fire than in unburned controls; however, mesophytic species such as red maple occur in much lower densities in fire-treated areas than in unburned controls (Gilbert et al. 2003).

Fires can also be utilized to help clear away dead or dying plant material, to increase production of native fire-tolerant species, promote environmental heterogeneity

and reduce the invasion of exotic plants (Anonymous 2000, Clark 1990). Fire may aid in

the control of plant diseases and decrease soil acidity. After a fire, shrubs sprout

prolifically and native herbs grow vigorously (Ahlgren and Ahlgren 1960).

Fire-dependent forest ecosystems require treatment with fire for their continued perpetuation on the landscape (Habeck and Mutch 1973). Fire suppression has created heavy fuel loads, high densities of old trees and an abundance of non-fire adapted species, leading to changes in the insects and diseases that act as agents of disturbance within a forest ecosystem (Parker et al. 2006). The ultimate result of fire suppression is the loss or deterioration of many ecosystem types and an increased risk of dangerous and intense wildfires.

FIRE-INSECT INTERACTIONS

Fire and insects are critical and intrinsic components of forest ecosystems that have the ability to alter and influence overall species composition, nutrient cycling and numerous other ecological processes. Fire can affect insects by killing them directly, altering soil properties, modifying overstory or understory vegetation structure or diversity, or modifying other aspects of habitat (McCullough et al. 1998). Insect outbreaks can also affect the hazard and severity of forest fires by altering the accumulation and distribution of fuels and vegetation. The interaction of fire and insects may delay or redirect forest succession and have significant consequences for productivity and biodiversity (McCullough et al. 1998).

Fire has been used by entomologists since the nineteenth century as an insect management tool (Miller 1979). Fire was applied to stands of eucalyptus to control

Phasmatidae, and to kill Acrididae on a variety of hosts (Miller 1979). Prescribed fire has been used to remove cerambycid-infested trees, and fire-killed larvae of cerambycids have been reported in lodgepole pine (Miller 1979). However, insect infestation can also be amplified by fire; insect infestation was positively related to the percentage of stand basal circumference killed by fire in Douglas fir and lodgepole pine

(Rasmussen et al. 1996). Firefighters have reported swarming cerambycids, buprestids and siricids after forest fires (Hermann et al. 1998). Factors influencing how fire affects insects include weather conditions, insect location and developmental stage, and fire intensity, seasonality, spread rate and coverage (Miller 1979, Wikars and Schimmel

2001).

10 When fire is viewed as an ecosystem process, rather than an insect management tool, its effects on arthropod populations are often not well-documented. Conservation and management strategies are often designed primarily from the point of view of vascular plants, rather than insects (Swengel 2001). The month, locality, soil type, past fire history and ground cover conditions should be included in the description of a prescribed burn in order to make informed decisions about the effects of a fire on arthropod populations (Hermann et al. 1998). Most insects decline sharply in abundance in the immediate hours and proceeding two months following a fire based on pre-fire and control sampling. The degree of arthropod population reduction may relate to the degree of direct exposure to flames (Swengel 2001). It is nearly impossible, though, to delimit effects from direct flame and smoke mortality from those that are a result of habitat alteration (Wikars and Schimmel 2001).

Fire seasonality has an effect on some insects. In the southeastern United States savanna ecosystems, many arthropods survive and recolonize best after a fire if it occurs during the growing season, when insects are most active (Hermann et al. 1998). The intensity of a fire may also alter its effect on insects. Fire caused an immediate change in beetle species composition, abundance and distribution in a Swedish pine forest: This pattern of change was stronger in areas that had experienced high intensity fires than in areas that had experienced low to moderate intensity fires. After 60 days this pattern became even more pronounced with a few pyrophilus species dominating the areas that had been subjected to high intensity fires (Wikars and Schimmel 2001).

Fire may also have indirect effects on arthropod populations by changing plant species composition and foliar characteristics, reducing litter layer and modifying soil moisture and temperature (Coleman and Rieske 2006). Fire induces changes in oak

11 foliar chemistry such as increases in foliar nitrogen, carbohydrates and defensive compounds, which may result in deterred herbivory by arthropods (Rieske-Kinney

2006). Burning reduces total microbial and fungal biomass, resulting in bottom-up regulation of soil litter arthropod communities (Coleman and Rieske 2006). Fire may also increase the quantity of coarse woody debris in an area, creating habitat for some arthropod species. Fire may benefit some arthropods by reducing competitors and

predators and increasing scavenger prey (Coleman and Rieske 2006).

Some insect species are fire-adapted and may even be fire-dependent.

Arthropods have adapted to fire by evolving morphological and behavioral characteristics that prevent mortality (Coleman and Rieske 2006). Invertebrates that are highly mobile and have thick cuticles are more likely to survive a fire than those which are less mobile and have thinner cuticles (Wikars and Schimmel 2001). There are nearly

40 documented European species of arthropods, mostly beetles that are fire-dependent.

These species declined during the twentieth century in large part due to fire suppression

(Hermann et al. 1998). Buprestid beetles of the genus Melanophilia have infrared receptors on their legs to orient themselves directly to radiant heat from fire (Miller

1979). Abundance of pollinators is correlated with decreased tree basal areas and increased percentage of herbaceous cover associated with prescribed fire in Georgia

(Campbell et al. 2007).

The reported effects of prescribed fire on ants (Hymenoptera: Formicidae) are

highly variable by habitat, ecosystem, species, fire seasonality and fire frequency.

Studies report that fire has no affect on formicids (Coleman and Rieske 2006). Other

studies report that ants are more abundant and diverse on recently burned sites (Miller

12 1979, Wilkinson et al. 2005). Still other studies find that some ant species increase in abundance following fire while others are reduced (Hanula and Wade 2003).

Desert harvester ant colonies (Pogonomyrmex spp.) can survive the direct effects of fire; however, fire can disrupt the foraging ecology of the ants by modification of the vegetation architecture, alteration of soil temperature and removal of shade vegetation

(Zimmer and Parmenter 1998). Ant species richness within forest leaf litter is similar across burned, burned and logged, unburned and unburned and logged habitats

(Andrew et al. 2000). Slashing and burning produced a shift from an omnivorous and nectivorous ant community to a wholly omnivorous ant community in agricultural crops (Castano-Meneses and Palacios-Vargas 2003). In longleaf pine, the biomass of ants is higher after dormant season fire than after growing season fire (New and Hanula

1998). In northern Arizona’s ponderosa pine forest, ant feeding guilds shifted after wildfires. Generalists were up to 25% more abundant after a wildfire than in a thinned or control stand. Opportunistic feeders were 34% less abundant in stands that had experienced wildfire than those that were thinned (Stephens and Wagner 2006).

Carpenter ant populations might be expected to increase after a fire, since they require logs or trees with at least some level of decay (which would assumedly be more prevalent after a fire) for colony nests (Leatherman 2002). However, a wide range of fire effects have been reported for carpenter ants, and it is likely that the effects of fire on these ants vary by habitat. Carpenter ant species richness was highest in control areas, followed by areas that had been burned at four year intervals, biennially, and annually, respectively, in a southeastern longleaf pine ecosystem (Hanula and Wade 2003). No differences have been found in Camponotus pennsylvanicus DeGeer abundance between once-burned and unburned areas in New Jersey pine barrens (Buffington 1967).

13 Other disturbances, such as thinning and deforestation, also have effects on ant diversity and abundance. Primary forests have higher species densities and higher ant diversity than secondary forests (Schonberg et al. 2004). Nest abandonment rates and local population sizes in a wood ant species, Formica aquilonia Yarrow, are much higher

in clear-cut forests than in uncut controls. In deforested areas, 39% of pre-deforestation

nests were abandoned, whereas, in an undisturbed control, less than 2% of nests were

abandoned (Sorvari and Hakkarainen 2007). While the exact effects may vary by ant

species, habitat type and fire effects, a strong collection of evidence exists to support the theory that disturbance is important in determining the composition and structure of ant communities (Andrew et al. 2000).

CARPENTER ANTS

Ants are often cited as the most ubiquitous creatures on the face of the earth

(Holldobler and Wilson 1991). They inhabit a myriad of ecosystems and are only naturally absent in Iceland, Greenland, and Antarctica (Folgarait 1998). Worldwide there are 15000 species in 296 genera (Folgarait 1998). All ants have elbowed antennae, an abdomen with the first segment fused to the thorax and the second segment elongated into a petiole, and division into morphologically distinct castes (Holldobler and Wilson 1991).

Carpenter ants belong to the subfamily Formicinae which is characterized by a one segmented petiole and a circular terminal cloacal orifice, usually with a fringe of hairs forming the acidopore (Warren and Rouse. 1969). Carpenter ants make up the genus Camponotus which comprises approximately 900 known species with 50 occurring

14 in North America. Camponotus workers are characterized by an unevenly convex

mesosomal dorsum and antennal scapes that insert behind the posterior edge of the

clypeus (Warren and Rouse 1969).

As their name implies, carpenter ants make their nests in many types of wood,

including structural, fallen trees, coarse woody debris, and standing trees. Carpenter

ant nests are divided into castes of queens, reproductive alates, major workers, minor

workers, various stages of larvae, and pupae. The black carpenter ant, C.

pennsylvanicus, is a large black ant recognized by its evenly rounded thorax, well marked

clypeal fossae, 12 segmented antennae, lack of sting, and antennal scapes that are never

flattened at the base. (Warren and Rouse 1969) Of the nearctic carpenter ant species, C.

pennsylvanicus is the largest and most widespread (Fowler and Roberts 1980).

Mature carpenter ant nests average 3000 individuals; the highest number of individuals recorded within a single nest was 8000 and was found in a C. pennsylvanicus colony (Hansen and Klotz 2005). In Indiana, black carpenter ants nest in standing trees and have an average foraging radius of 1-6 trees (Hansen and Klotz 2005). When nests become too large, satellite colonies are established vertically within the bole of the tree and horizontally on the forest floor. The original nest site is known as the parent colony.

Parent colonies are usually confined to the basal two meters of the bole of standing trees

(Hansen and Klotz 2005). If conditions within the parent colony become unsuitable, the nest site will be abandoned and a new colony will be established. If conditions remain favorable, a colony can persist from 7-21 years, or the lifespan of a queen (Hansen and

Klotz 2005). In areas where C. laevigatus Smith and C. floridanus Buckley coincide with C. pennsylvanicus, the nests of C. pennsylvanicus are found nearly exclusively in standing

15 timber, whereas the nests of the other two species are found in woody debris along the forest floor (Klotz et al. 1998).

Major workers perform the majority of foraging, whereas minor workers maintain the nest. Nutrients are transferred among members of the colony via trophollaxis, the transfer of foods among members of the community from mouth to mouth (Hansen and Klotz 2005). Carpenter ants preferentially collect nitrogen and carbohydrate based food sources (Cannon 1998). They tend honeydew-producing

Homoptera such as aphids which provide sugary substances in exchange for defense; the ants also will scavenge and prey upon other insects (Hansen and Klotz 2005). While carpenter ants are facultatively predaceous, during insect outbreaks, other insects can become up to 90% of the ants’ diet (Youngs and Campbell 1984). In times of resource shortage, carpenter ant larvae are facultatively cannibalized by the workers of the colony. Resources are often partitioned with Formica subserica Say in areas where their ranges overlap. Black carpenter ants forage most heavily at night, whereas F. subserica forages during daylight hours.

Foragers make up less than 10% of the colony’s individuals. The rest of the individuals are devoted to grooming, nest maintenance, larval care and defense of territory. Foraging activity is high from May through September, peaking in mid-July; this peak corresponds to maturation of the larval brood (Sanders 1972). Night length was positively correlated with foraging activity, and wind speed was negatively correlated with foraging activity by C. pennsylvanicus (Nuss et al. 2005).

In the spring of the year, winged male alates leave the nest and release mandibular gland chemicals that stimulate the winged female alates to follow. The alates then swarm and engage in mating flight. After mating, each newly fertilized

16 female removes her wings and begins seeking a suitable nest site; males die shortly after inseminating females (Hansen and Klotz 2005). Galleries of wood-boring insects, such as cerambycid beetles, are frequently chosen as starting sites for nest initiation. Upon identification of a nest site, the fertilized female alate, a new queen, lays her first brood of 9-16 eggs. These eggs hatch in 2-5 weeks. The larvae go through a series of four instars, in approximately 2-3 weeks. Pupae develop into adults in 2-4 weeks. The entire cycle takes from 48-74 days to complete (Hansen and Klotz 2005). The queen tends the first brood of larvae herself. Subsequent broods of young are cared for by workers.

With the exception of the first year when a single cohort is produced, two cohorts of young occur each year. A newly established nest will typically have 4-25 workers after one year. Colonies do not produce alates until 3-6 years after initiation (Hansen and

Klotz 2005).

Carpenter ants make up 97% of the diet of the pileated woodpecker, Dryocopus

pileatus L. (Picidae) (Beckwith and Bull 1985). They also serve as food sources for other

birds. Phorid and chalcid flies parasitize the larvae and pupae of C. pennsylvanicus,

respectively. Myrmecophile insects such as Xenodusa (Staphylnidae), beetles in the

family Leiodidae and small crickets have been found living in nests of C. pennsylvanicus

(Hansen and Klotz 2005). Two fungi, Desmidiospora myrmecophila and Cordyceps species, have been documented as pathogens of black carpenter ants (Mains 1958, Chen et al.

2002).

17 ANT PREDATION

Ants occupy a variety of ecological roles, acting as scavengers, herbivores,

fungivores, detritivores and predators. Many ant species are facultative predators,

including C. pennsylvanicus. Ants will prey upon other ant species and engage in

intraspecific egg, larval and pupal cannibalism (Risch and Carroll 1982). Ants continue

to forage long after their physiological capacity for nutrients has been met. Predator

satiation does not limit ants’ ability to collect prey items, because they can store food in

their colonies (Risch and Carroll 1982). Ant predation rates are higher in tropical areas

than in temperate areas (Jeanne 1979). Urban areas exhibit lower rates of predation than

natural ecosystems at all latitudes (Jeanne 1979).

While the role of ants as predators of forest pest insects is not fully understood, strong evidence exists to support the idea that ants are potentially important generalist natural enemies of forest pests. Ants have been observed as predators of western spruce budworm during outbreaks (Young and Campbell 1984). When ants were excluded from trees with western spruce budworm pupae, 8% of the pupae were removed, whereas trees where ants were not excluded had 85% of pupae removed (Campbell and

Torgersen 1982). In a separate study, three species of Camponotus and six Formica species removed 70-75% of western spruce budworm pupae from sample trees within four days

(Youngs and Campbell 1984). Carpenter ants have also been observed as predators of jackpine budworm and the Swaine jackpine sawfly (Smirnoff 1959, Allen et al. 1970).

C. pennsylvanicus and Aphaenogaster tennessensis Mayr are confirmed predators of

red oak borer, Enaphalodes rufulus (Haldeman) (Muilenburg et al. 2008). Both ant species were observed removing artificially planted red oak borer eggs from northern red oaks,

18 Quercus rubra at the rate of 70% in the first hour. Black carpenter ants removed the majority of these eggs. Ants were also seen carrying away small red oak borer larvae.

The remains of red oak borers were also identified in the gut of wild-collected black carpenter ant workers using molecular detection techniques (Muilenburg et al. 2008).

Ants may make up as much as 13-29% of the total predation occurring on red oak borer at endemic population levels; however, the importance of ants as predators of the red oak borer at epidemic population levels is still unclear (Galford 1985).

RED OAK BORER

Red oak borer is a native wood-boring beetle (Cerambycidae: Coleoptera) in eastern North America. The insect is nearly ubiquitous in eastern hardwood forests, and historically has been recorded as a minor pest of oak timber, reducing lumber value, breeding in living trees but rarely causing tree mortality (Donley 1983).

Red oak borer has a two year life cycle with synchronous adult emergence occurring only in odd-numbered years. Adults emerge in June and July, do not feed and are nocturnal (Fierke et al. 2005, Galford 1974). Adult borers live an average of 19-22 days (Donley 1978). Females oviposit an average of 119 eggs in bark crevices and under vines and lichens (Fierke et al. 2005). Up to 95% of the eggs hatch (Donley and Acciavatti

1980). Larvae bore directly through the outer bark then spend four months, until mid-

November, actively feeding in phloem tissue. First year larvae overwinter in the phloem until late spring. In the second year, larvae feed in the phloem and then create galleries within the sapwood. Second year larvae overwinter in the sapwood from November until June, with galleries protected by a frass plug. Pupation occurs in May (Fierke et al.

19 2005). External evidence of red oak borer infestation includes emergence holes, attack holes, extruded frass, discolored patches of bark and wood slivers (Donley and

Acciavatti 1980).

The majority of red oak borer attacks (99%) occur on the bole of the tree (Donley and Rast 1984). Tree age is significantly correlated with red oak borer attack, with older trees receiving greater numbers of attacks than younger trees (Fierke et al. 2005).

While drought may play a role in general oak decline, no relation between precipitation and red oak borer abundance has been observed. (Muzika and Guyette 2004).

Red oak borer can develop in northern red oak, scarlet oak, black oak and white oak (Fierke et al. 2005); however, white oak is not a preferred host (Galford 1983). Red oak borer prefers black oak over scarlet and northern red oak, but all three tree species exhibit similar larval mortality rates, with the highest mortality occurring in the first year of larval development (Hay 1974, Galford 1985).

Galford (1985) reported that on average 15% of individual red oak borers survive

to adulthood; however, survival to adulthood is four times greater when individual

larvae are protected from predation. Ants and woodpeckers account for a combined 40-

60% of larval mortality (Galford 1985). Other predators include carpenterworms and

elaterid larvae; however, nitidulids exhibit no predaceous behavior (Ware and Stephen

2006). Brentid, cossid and elaterid larvae (Coleoptera), ants and decay organisms utilize

red oak borer attack holes and galleries to gain entrance into the heartwood of oaks

(Donley and Acciavatti 1980).

The 1999 Ozark red oak decline event coincided with an unprecedented explosion in red oak borer populations. Over half (55%) of the red oaks in the Ozark

National Forest are in heavily infested stands (Stephen et al. 2001). Pre-epidemic studies

20 estimated borer populations at less than one individual per tree, whereas during the epidemic a mean of 77 borer larvae were found per tree, with a range of 0-577 (Fierke et al. 2005). As many as 1244 current red oak borer galleries were found in a single tree, and 5041 attack holes were documented on an individual tree (Fierke et al. 2005).

Populations returned to endemic levels by 2005; however, 13% of the 527 red oaks monitored annually during the outbreak died and 31% declined (Fierke et al. 2007).

RESEARCH OBJECTIVES

To date, there is no satisfactory explanation for the red oak borer population outbreak and crash. One way populations are regulated and controlled is via predation.

A sharp decrease or increase in predation would theoretically allow a population to rapidly increase or decrease, respectively, in a relatively short period of time. Since black carpenter ants are known predators of red oak borer eggs, the relationship

between black carpenter ant and red oak borer distribution may provide insight into the

factors influencing the red oak borer population explosion and crash. As prescribed fire

becomes an increasingly prevalent management tool in oak hickory forests, it is

important to comprehensively evaluate its effects on insect populations, particularly

those that may be potentially significant predators of forest pests.

My general objectives are to (1) examine differences in black carpenter ant abundance as they vary by red oak borer infestation levels, (2) examine site and stand variables that may influence black carpenter ant distribution, and (3) examine the effects of prescribed fire on black carpenter ant abundance. To accomplish these objectives, I assess foraging

21 preferences of black carpenter ants on red oaks and correlate these preferences to previous numbers of red oak borer emergence holes, site aspect and tree class. I also evaluate the effects of dormant season prescribed fire, tree species, and tree size on black carpenter ant abundance and distribution in the oak-hickory forests of the Ozark

Mountains of Arkansas.

22 REFERENCES

Ahlgren, I.F. and C.E. Ahlgren. 1960. Ecological effects of forest fires. 1960. Botanical Review 483-517.

Allen, D.C., F.B. Knight and J.L. Foltz. 1970. Invertebrate predators of the jack-pine budworm, Choristoneura pinus, in Michigan. Annals of the Entomological Society of America 63: 59-64.

Andrew, N., L. Rodgerson and A. York. 2000. Frequent fuel-reduction burning: the role of logs and associated leaf litter in the conservation of ant biodiversity. Australian Journal of Ecology 25: 99-107.

Anonymous. 2000. Fire ecology in the southeastern United States. Technical Report USGS-FS-018-00. United States Geological Survey. Lafayette, LA. 4 p.

Anonymous. 2007. BLM Wyoming wildland fire glossary. United States Department of the Interior Bureau of Land Management. http://www.blm.gov/wy/st/en/programs/Fire/glossary.html. 28 December 2007.

Anonymous. 2008. Ozark St. Francis National Forests. USDA Forest Service. http://www.fs.fed.us/oonf/ozark/. 10 February 2008.

Batek, M.J., A.J. Rebertus, W.A. Schroeder, T.L. Haithcoad, E. Compas and R.P. Guyette. 1999. Reconstruction of early nineteenth-century vegetation and fire regimes in the Missouri Ozarks. Journal of Biogeography 26: 397-412.

Beckwith, R.C. and E.L. Bull. 1985. Scat analysis of the arthropod component of the pileated woodpecker diet. The Murrelet 66: 90-92.

Beilmann, A.P. and L.G. Brenner. 1951. The changing forest flora of the Ozarks. Annals of the Missouri Botanical Garden 38: 283-291.

Brennan, L.A. and S.M. Hermann. 1994. Prescribed fire and forest pests: Solutions for today and tomorrow. Journal of Forestry 92: 34-37.

Brockway, D.G. and C.E. Lewis. 1997. Long-term effects of dormant-season prescribed fire on plant community diversity, structure and productivity in a longleaf pine wiregrass ecosystem. Forest Ecology and Management 96: 167-183.

Brose, P. 2004. Fitting fire into oak management. In Upland Oak Ecology Symposium. Gen. Tech. Rep. SRS-73. Asheville, NC: USDA, M.A. Spetich ed. Forest Service Southern Research Station. 263.

23 Brose, P. and D. VanLear. 2004. Survival of hardwood regeneration during prescribed fires: the importance of root development and root collar location. In Upland Oak Ecology Symposium. Gen. Tech. Rep. SRS-73. Asheville, NC: USDA, M.A. Spetich ed. Forest Service Southern Research Station. 123-127.

Brose, P., D. VanLear and R. Cooper. 1999. Using shelterwood harvests and prescribed fire to regenerate oak stands on productive upland sites. Forest Ecology and Management 113: 125-141.

Buffington, J.D. 1967. Soil arthropod populations of the New Jersey pine barrens as affected by fire. Annals of the Entomological Society of America 60: 530-535.

Campbell, J.W., J.L. Hanula and T.A. Waldrop. 2007. Effects of prescribed fire and fire surrogates on floral visiting insects of the blue ridge province in North Carolina. Biological Conservation 134: 393-404.

Campbell, R.W. and T.R. Torgersen. 1982. Some effects of predaceous ants on western spruce budworm in north central Washington. Environmental Entomology 11: 111-114.

Cannon, C.A. 1998. Nutritional ecology of the carpenter ant Camponotus pennsylvanicus (DeGeer): Macronutrient Preference and Particle Consumption [dissertation]. Blacksburg, Virginia. Virginia Polytechnic Institute and State University.

Castano-Meneses, G. and J.G. Palacios-Vargas. 2003. Effects of fire and agricultural practices on neotropical ant communities. Biodiversity and Conservation 12: 1913-1919.

Chen, Y., L.D. Hansen and J.J. Brown. 2002. Nesting sites of the carpenter ant, Camponotus vicinus (Mayr) (Hymenoptera: Formicidae) in Northern Idaho. Environmental Entomology 31: 1037-1042.

Clark, J.S. 1990. Twentieth-century climate change, fire suppression, and forest production and decomposition in northwestern Minnesota. Canadian Journal of Forestry Research 20: 219-232.

Clements, F.E. 1920. Plant indicators: the relation of plant communities to process and practice. Carnegie Institute Publishing. Washington, D.C. No. 290. 453 p.

Coleman, T.W. and L.K. Rieske. 2006. Arthropod response to prescription burning at the soil-litter interface in oak-pine forests. Forest Ecology and Management 233: 52-60.

Countryman, C.M. 1972. The fire environment concept. USDA Forest Service Pacific Southwest Forest and Range Experiment Station. Berkley, CA. 12 p.

24 Dey, D.C. and G. Hartman. 2005. Returning fire to Ozark Highland forest ecosystems: effects on advanced regeneration. Forest Ecology and Management 217: 37-53.

Donley, D.E. 1978. Oviposition by the red oak borer, Enaphalodes rufulus Coleoptera: Cerambycidae. Annals of the Entomological Society of America 71: 496-498.

Donley, D.E. 1983. Cultural control of the red oak borer (Coleoptera: Cerambycidae) in forest management units. Journal of Economic Entomology 76: 927-929.

Donley, D.E. and R.E. Acciavatti. 1980. Red oak borer. USDA Forest Service. Forest Insect and Disease Leaflet 163. 5 p.

Donley, D.E. and E. Rast. 1984. Vertical distribution of red oak borer, Enaphalodes rufulus (Coleoptera: Cerambycidae) in red oak. Environmental Entomology 13: 41-44.

Fierke, M.K., D.L. Kinney, V.B. Salisbury, D.J. Crook and F.M. Stephen. 2005. Development and comparison of intensive and extensive sampling methods and preliminary within-tree population estimates of red oak borer (Coleoptera: Cerambycidae) in the Ozark Mountains of Arkansas. Environmental Entomology 34: 185-192.

Fierke, M.K., M.B. Kelley and F.M. Stephen. 2007. Site and stand variables influencing red oak borer, Enaphalodes rufulus (Coleoptera: Cerambycidae), population densities and tree mortality. Forest Ecology and Management 247: 227-236.

Folgarait, P.J. 1998. Ant biodiversity and its relationship to ecosystem functioning: a review. Biodiversity and Conservation 7: 1221-1244.

Foti, T.L. 2004. Upland hardwood forests and related communities of the Arkansas Ozarks in the early 19th century. In Upland Oak Ecology Symposium. Gen. Tech. Rep. SRS-73. Asheville, NC: USDA, M.A. Spetich ed. Forest Service Southern Research Station. 21-29.

Fowler, H.G. and R.B. Roberts. 1980. Foraging behavior of the carpenter ant, Camponotus pennsylvanicus (Hymenoptera: Formicidae) in New Jersey. Journal of the Kansas Entomological Society 53: 295-304.

Fralish, J.S. 2004. The keystone role of oak and hickory in the central hardwood forest. In Upland Oak Ecology Symposium. Gen. Tech. Rep. SRS-73. Asheville, NC: USDA, M.A. Spetich ed. Forest Service Southern Research Station. 78-87.

Galford, J.R. 1974. Some physiological effects of temperature on artificially reared red oak borers. Journal of Economic Entomology 67: 709-710.

25 Galford, J.R. 1983. Life history of the red oak borer, Enaphalodes rufulus (Haldeman) in white oak (Coleoptera: Cerambycidae). Entomological News 94: 7-10

Galford, J.R. 1985. Role of predators on an artificially planted red oak borer population. USDA Forest Service Northeastern Forest Experiment Station. Research Note NE-331. 2 p.

Gilbert, N.L., S.L. Johnson, S.K. Gleeson, B.A. Blankenship and M.A. Arthur. 2003. Effects of prescribed fire on physiology and growth of Acer rubrum and Quercus spp. seedlings in an oak-pine forest on the Cumberland Plateau, KY. Journal of the Torrey Botanical Society 130: 253-264.

Guyette, R.P. and M.A. Spetich. 2003. Fire history of oak-pine forests in the Lower Boston Mountains, Arkansas, USA. Forest Ecology and Management 180: 463- 474.

Habeck, J.R. and R.W. Mutch. 1973. Fire-dependent forests in the northern Rocky Mountains. Quaternary Research 3: 408-424.

Hansen, L.D. and J.H. Klotz. 2005. Carpenter ants of the United States and Canada. Ithaca, NY. Cornell University Press. 224 p.

Hanula, J.L. and D.D. Wade. 2003. Influence of long-term dormant-season burning and fire exclusion on ground-dwelling arthropod populations in longleaf pine flatwoods ecosystems. Forest Ecology and Management 175: 163-184.

Hay, C.J. 1974. Survival and mortality of red oak borer larvae on black, scarlet, and northern red oak in eastern Kentucky. Annals of the Entomological Society of America 67: 981-986.

Heitzman, E. 2003. Effects of oak decline on species composition in a northern Arkansas forest. Southern Journal of Applied Forestry 27: 264-268.

Hermann, S.M., T. VanHook, R.W. Flowers, L.A. Brennan, J.S. Glitzenstein, D.R. Streng, J.L. Walker and R.L. Myers. 1998. Fire and biodiversity: studies of vegetation and arthropods. In Transactions of the 63rd North American Wildlife and Natural Resources Conference. Orlando, FL. Washington, DC Wildlife Management Institute. 384-401.

Holldobler, B. and E.O. Wilson. 1991. The Ants. Cambridge, MA. Harvard University Press. 756 p.

Jeanne, R.L. 1979. A latitudinal gradient in rates of ant predation. Ecology 60: 1211- 1224.

26 Jenkins, M.A. and S.G. Pallardy. 1995. The influence of drought on red oak group species growth and mortality in the Missouri Ozarks. Canadian Journal of Forestry Research 25: 1119-1127.

Jurney, D., R. Evans, J. Ippolito and V. Bergstrom. 2004. The role of wildland fire in portions of southeastern North America. In Proceedings of the 22nd Tall Timbers Fire Ecology Conference: Fire in Montane, Boreal, and Temperate Ecosystems. Tallahassee, Florida: R.T. Engstrom, K.E.M. Gallery, and W.J. deGroot eds. 95- 116.

Klotz, J.H., L. Greenberg, B.L. Reid and L. Davis Jr. 1998. Spatial distribution of colonies of three carpenter ants, Camponotus pennsylvanicus, Camponotus floridanus, Camponotus laevigatus (Hymenoptera: Formicidae). Sociobiology 32: 51-62.

Leatherman, D. 2002. Insects and diseases associated with forest fires. Colorado State University Cooperative Extension. FS-6309. Boulder, CO. 3 p.

Leopold, A. 1949. A Sand County Almanac. The Oxford University Press. New York, NY. 295 p.

Loewenstein, E.F., P.S. Johnson and H.E. Garrett. 2000. Age and diameter structure of a managed uneven-aged oak forest. Canadian Journal of Forest Research 30: 1060-1070.

Mains, E.B., 1958. North American entomogenous species of Cordyceps. Mycologia 50: 169-222.

Manion, P.D. and D. LaChance. 1992. Forest decline concepts: An overview. In: Manion, P.D., Lachance, D. (Eds), Forest Decline Concepts. Saint Paul, MN. American Phytopathological Society Press. 181-190.

McCullough, D.G., R.A. Werner and D. Neumann. 1998. Fire and insects in northern and boreal forest ecosystems of North America. Annual Review of Entomology 43: 107-127.

Miller, W.E. 1979. Fire as an insect management tool. Entomological Society of America Bulletin. 25: 137-140.

Mueller-Dombois, D. 1992. A natural dieback theory, cohort senescence as an alternative to the decline disease theory. In Forest Decline Concepts Manion, P.D. and D. Lachance eds. American Phytopathological Society Press, Saint Paul, MN. 26-37.

Muilenburg, V.L., F.L. Goggin, S.L. Hebert, L. Jia and F.M. Stephen. 2008. Ant predation on red oak borer confirmed by field observation and DNA analysis. Agricultural and Forest Entomology 10: 1-9.

Muzika, R.M. and R.P. Guyette. 2004. A dendrochronological analysis of red oak borer abundance. In Upland Oak Ecology Symposium. Gen. Tech. Rep. SRS-73. Asheville, NC: M.A. Spetich ed. Forest Service Southern Research Station. 102- 105.

New, K.C. and J.L. Hanula. 1998. Effects of time elapsed after prescribed burning in longleaf pine stands on potential prey of the red-cockaded woodpecker. Southern Journal of Applied Forestry 22: 175-183.

National Oceanic and Atmospheric Administration (NOAA). 2007. United States Department of Commerce. World Wide Web. http://www.noaa.gov. 12 November 2007.

Nuss, A.B., D.R. Suiter and G.W. Bennett. 2005. Continuous monitoring of the black carpenter ant, Camponotus pennsylvanicus (Hymenoptera: Formicidae), trail behavior. Sociobiology 45: 597-618.

Oak, S.W., C.M. Huber and R.M. Sheffield. 1986. Incidence and impact of oak decline in western Virginia, 1986. USDA Forest Service Resource Bulletin SE-123. Southeastern Forest Experiment Station. Asheville, NC. 19 p.

Ozark-St Francis National Forests. 1978. Timber management plan, final environmental statement. USDA-FS-R8 Admin 77-07. USDA Forest Service. 113 p.

Palmer, E.J. 1921. The forest flora of the Ozark Region. Journal of the Arnold Arboretum 2: 216-232.

Parker, T.J., K.M. Clancy and R.L. Mathiasen. 2006. Interactions among fire, insects and pathogens in coniferous forests of the interior western United States and Canada. Agricultural and Forest Entomology 8: 167-189.

Pedersen, B.S. 1998. The role of stress in the mortality of Midwestern oaks as indicated by growth prior to death. Ecology 79: 79-93.

Rasmussen, L.A., G.D. Amman, J.C. Vandygriff, R.D. Oakes, A.S. Munson and K.E. Gibson. 1996. Bark beetle and wood borer infestations in the greater Yellowstone area during four postfire years. Forest Service Intermountain Research Station. INT-RP-487. 10 p.

Read, R.A. 1952. Tree species occurrence as influenced by geology and soil on an Ozark north slope. Ecology 33: 239-246

Rieske-Kinney, L.K. 2006. Do fire and insects interact in the eastern forest? In Fire in the Eastern Forest: Delivering Science to Land Managers. Gen. Tech. Rep. NRS-

28 P-1. Columbus, OH: USDA, M.B. Dickinson ed. Forest Service Northern Research Station. 152-157.

Risch, S.J. and C.R. Carroll. 1982. The ecological role of ants in two Mexican agroecosystems. Oecologia 55: 114-119.

Sanders, C.J. 1964. The biology of carpenter ants in New Brunswick. Canadian Entomologist 96: 894-909.

Sanders, C.J. 1972. Seasonal and daily activity patterns of carpenter ants (Camponotus spp.) in Northwestern Ontario (Hymenoptera: Formicidae). Canadian Entomologist 104: 1681-1687.

Saveland, J. 1995. Fire in the forest. In Forest Health Through Silviculture: Proceedings of the 1995 National Silviculture Workshop. Gen. Tech. Rep. RM-GTR-267. Mescalero, NM: USDA, L.G. Eskew, ed. Forest Service Rocky Mountain Research Station. 14-19.

Schonberg, L.A., J.T. Longino, J.T. Nadkarni and S.P. Yanoviak. 2004. Arboreal ant species richness in primary forest, secondary forest, and pasture habitats of a tropical montane landscape. Biotropica 36: 402-409.

Smirnoff, W.A. 1959. Predators of Neodiprion swainei Midd. (Hymenoptera: Tenthredinidae) larval vectors of virus diseases. Canadian Entomologist 91: 246-248.

Smith, D.W. 2006. Why sustain oak forests? In Fire in the Eastern Forest: Delivering Science to Land Managers. Gen. Tech. Rep. NRS-P-1. Columbus, OH: USDA, M.B. Dickinson ed. Forest Service Northern Research Station. 152-157.

Sorvari, J. and H. Hakkarainen. 2007. Wood ants are wood ants: deforestation causes population declines in the polydomous wood ant Formica aquilonia. Ecological Entomology 32: 707-711.

Spetich, M.A. 2004. Early changes in physical tree characteristics during an oak decline event in the Ozark Highlands. In Upland Oak Ecology Symposium. Gen. Tech. Rep. SRS-73. Asheville, NC: USDA, M.A. Spetich ed. Forest Service Southern Research Station 424-427.

Starkey, D.A., A. Mangini, F.L. Oliveria, S.R. Clark, B. Bruce, R. Kertz and R. Menard. 2000. Forest health evaluation of oak mortality and decline on the Ozark National Forest, 1999. Forest Health Protection Report 2000-0-02. 31 p.

Stephen, F.M., V.B. Salisbury and F.L. Oliveria. 2001. Red oak borer, Enaphalodes rufulus (Coleoptera: Cerambycidae), in the Ozark Mountains of Arkansas, U.S.A.: An unexpected and remarkable forest disturbance. Integrated Pest Management Reviews 6: 247-252.

Stephens, S.S. and M.R. Wagner. 2006. Using ground foraging ant (Hymenoptera: Formicidae) functional groups as bioindicators of forest health in northern Arizona ponderosa pine forests. Environmental Entomology 35: 937-949.

Swengel, A.B. 2001. A literature review of insect responses to fire, compared to other conservation managements of open habitat. Biodiversity and Conservation 10: 1141-1169.

VanLear, D.H. 2004. Upland oak ecology and management. In Upland Oak Ecology Symposium. Gen. Tech. Rep. SRS-73. Asheville, NC: USDA, M.A. Spetich ed. Forest Service Southern Research Station. 65-71.

Ware, V.L. and F.M. Stephen. 2006. Facultative intraguild predation of red oak borer larvae (Coleoptera: Cerambycidae). Environmental Entomology 35: 443-447.

Wargo, P.M., D.R. Houston and L.A. LaMadeleine. 1983. USDA Forest Service. Forest Insect and Disease Leaflet 16. 7 p.

Warren, L.O. and E.P. Rouse. 1969. The ants of Arkansas. Agricultural Experiment Station. University of Arkansas, Fayetteville. Bulletin 742. 69 p.

Wikars, L.O. and J. Schimmel. 2001. Immediate effects of fire-severity on soil invertebrates in cut and uncut pine forests. Forest Ecology and Management 141: 189-200.

Wilkinson, E.B., E.G. Lebrum, M.L. Spencer, C. Whitby and C. Kleine. 2005. Short- term effects of fire on Sky Island ant communities. In USDA Forest Service Proceedings RMRS-P-36. Boulder, CO: USDA Forest Service Rocky Mountain Research Station. 550-552.

Youngs, L.C. and R.W. Campbell. 1984. Ants preying on pupae of the western spruce budworm Choristoneura occidentalis (Lepidoptera: Tortricidae) in eastern Oregon and western Montana. Canadian Entomologist 116: 1665-1669.

Zimmer, K. and R.R. Parmenter. 1998. Harvester ants and fire in a desert grassland: ecological responses of Pogonomyrmex rugosus (Hymenoptera: Formicidae) to experimental wildfires in Central New Mexico. Environmental Entomology 27: 282-287.

CHAPTER 2

OCCURRENCE OF BLACK CARPENTER ANTS IN TREES PREVIOUSLY INFESTED WITH RED OAK BORER

31 INTRODUCTION

The red oak borer (Enaphalodes rufulus Haldeman) is a native wood-boring beetle that has been implicated as an agent of tree mortality in a large-scale oak decline event

(>300000 hectares) across the Ozark and Ouachita Mountains of Arkansas and Missouri

(Stephen et al. 2001, Heitzman 2003, Fierke et al. 2005). Historically, the red oak borer was not considered as a tree killer, and previous red oak borer studies document the insect at low populations; however, during the recent outbreak, over 5000 larval attacks were found on a single tree (Hay 1974, Fierke et al. 2005a). Populations remained high

from 1999 until 2005 (Fierke et al. 2007) but have since declined to endemic levels (J.

Riggins 2008).

Red oak borer has a two year synchronous life cycle with adults emerging in early summers of odd-numbered years (Donley and Acciavatti 1980). Adults live for approximately 3 weeks and do not feed; mating and oviposition occurs for approximately 16 days in mid-summer (Galford 1974, Fierke et al. 2005a). A female red oak borer lays an average of 120 eggs on tree boles in bark crevices and under lichens.

Eggs hatch within two weeks (Donley and Acciavatti 1980) and larvae bore into and consume phloem tissues until mid-November (Hay 1969). First year larvae then enter a state of quiescence and overwinter within the phloem gallery (Hay 1969). In late spring to early summer of the next year, larvae resume phloem feeding and eventually excavate a gallery into the heartwood (Hay 1969). Larvae overwinter the second year within the heartwood gallery protected by a frass plug. The beetles pupate in May of the second year and emerge shortly thereafter (Hay 1969).

32 Intraspecific predation, woodpeckers, carpenterworms and ants have been cited as agents of borer mortality (Hay 1972, Donley and Acciavatti 1980, Galford 1985, Ware and Stephen 2006). Ants may be responsible for 13-29% of the predation on red oak borer at endemic levels (Galford 1985). Camponotus pennsylvanicus DeGeer, the black carpenter ant, and Aphaenogaster tennessensis Mayr have been observed removing red oak borer eggs from tree surfaces at the rate of 70% in the first hour and red oak borer

DNA has been detected in the guts of wild-collected black carpenter ants (Muilenberg et al. 2008).

Evidence suggests that carpenter ants are important predators of forest pests

(Parmelee 1941, Allen et al. 1970, Smirnoff 1970, Youngs and Campbell 1984, Feicht 1985,

Galford 1985, Muilenberg 2008). Black carpenter ants may also have a role in controlling red oak borer populations at both epidemic and endemic levels (Galford 1985,

Muilenberg 2008). Red oak borer populations have been monitored since 2001 at three sites in the Ozark National Forest of Arkansas (Fierke et al. 2007). Rapid estimation procedure (REP) classes have also been developed to rapidly assess red oaks for red oak borer infestation. This scale uses the number of red oak borer emergence holes in the basal two meters of a tree and the percent dieback in the crown (Fierke et al. 2005a).

These data can be correlated to other variables on the same trees. The relationship between black carpenter ant and red oak borer distribution may provide insight into the factors influencing the red oak borer population explosion and decline. The objectives of this study were to correlate the presence of black carpenter ants on previously infested red oaks to previous numbers of red oak borer emergence holes, site aspect and tree class.

33 METHODS

Study Areas

Three geographic locations, Fly Gap, White Rock and Oark (UTM Zone 15-N

NAD83: Fly Gap: 0431660, 3954978; White Rock: 0412668, 3949429; Oark: 0450792,

3952369), within the Ozark National Forest were chosen for this study. Stands at these locations were 70-100 years in age and on shallow rocky limestone and sandstone soils

(USDA Forest Service 1993, Starkey et al. 2000). These three locations are in areas previously identified as experiencing high red oak mortality. Plots at each location, when originally established had an average of 493 trees per ha (Fierke et al. 2007). All plots were dominated by northern red oaks (Quercus rubra L.) prior to the red oak borer outbreak and are part of the oak-hickory forest type that is characteristic of the Ozark

Mountains (USDA Forest Service 1993, Fierke et al. 2007).

Each of these locations had a single pre-established 30x100 meter fixed area rectangular vegetation plot on each of north, south, east and west benches and ridges on which all northern red oaks have been annually monitored since 2001 for red oak borer attack holes, emergence holes, tree crown conditions and red oak mortality (Fierke et al.

2005a). A total of 15 plots were available for use in this study. This study was conducted using 13 of these 15 available plots. The ridge plot at White Rock had been highly impacted by tree removal and the west plot at Fly Gap had been treated with prescribed fire, so they were thus deemed unsuitable and excluded.

34 Field Experiments

In each plot, all red oaks larger than 10 cm diameter at breast height (1.3 m above the ground) were identified and marked. Trees were baited using 5 mL of a 1:1 mixture of clover honey and tuna in oil. Baits were placed in 10 mL paper cups and a single bait cup was affixed to each tree at breast height using aluminum nails. The bait was left undisturbed in the plot for one hour then ants conspicuously visible on the bole of the tree and in bait cups were collected and returned to the laboratory for identification. All trees were evaluated for ants on a presence/absence basis because the number of ants present on a tree gives little indication of the number of colonies or individuals per unit area (Kilpelainen et al. 2005). All trees were sampled from the hours of 06:00-11:00 because ants will not forage in extreme afternoon heat (Nuss et al. 2005). The presence of ants was analyzed in relation to aspect, geographic location, tree class and red oak borer emergence holes.

Data Analysis

This experiment was analyzed as a randomized incomplete block. The effect of aspect on black carpenter ant presence was analyzed using all red oaks. A one-way analysis of variance (ANOVA) was performed (α=0.05) and Tukey pairwise comparison tests were conducted to analyze significant differences when appropriate. The effects of geographic areas were also analyzed by one-way ANOVA (α=0.05) and Tukey comparisonwise tests to examine where statistically significant differences existed.

35 Tree classes were assigned using the rapid estimation procedure (REP) developed by Fierke et al. (2005a) on the basis of tree crown condition and the number of red oak borer emergence holes in the basal 2 m of the tree. Tree class categories represent red oak borer infestation history and ranged from I to III. Class I trees were healthiest; Class II trees had an intermediate level of health characterized by a declining crown and at least five red oak borer emergence holes. Class III trees were least healthy, with high numbers of red oak borer emergence holes and a severely declining crown

(Fierke et al. 2005b). Tree REP class was analyzed against mean black carpenter ant abundance using a one-way ANOVA (α=0.05). Tukey pairwise comparison tests were used to analyze differences.

The REP class consisted of two variables: crown condition and the number of red oak borer emergence holes on the basal 2 m of a tree. Since it was my objective to assess black carpenter ants in relation to red oak borer distribution, red oak borer emergence holes were analyzed independently. The numbers of red oak borer emergence holes that were visible in 2005 was analyzed by converting the number of holes to a class. Trees that had 0-5 visible red oak borer emergence holes were classified as “low.” Trees with >5 visible red oak borer emergence holes were classified as “high.”

The classes low and high were analyzed against black carpenter ant abundance using a one-way ANOVA and Tukey comparison-wise tests.

Data were summarized as percentages of trees of a given attribute (e.g. high/low levels of red oak borer emergence holes, situation on a given aspect, etcetera) with black carpenter ants present and normalized using an arc-sine transformation. Data were analyzed using JMP 7.0 (SAS Institute 2007). Voucher specimens were deposited in the

Insect Museum at the University of Arkansas, Fayetteville, Arkansas.

RESULTS

A total of 328 red oaks were surveyed for ants in this study. Stand characteristics

are reported in Table 1. An overall average of 77.1% red oaks had ants present

(SE=0.02). Aspect was not significantly correlated to black carpenter ant presence

(p=0.43, df=327). Geographic location was also not significantly correlated to black

carpenter ant presence (p=0.21, df=327).

Tree class was significantly correlated to black carpenter ant presence (p=0.003, df=327) (Figure 1). Ant presence on Class I and Class II trees was not significantly different; however, ants were present significantly less on Class III trees than either

Class I or Class II trees. Black carpenter ants were found on 81% of Class I trees

(SE=0.04, N=126), 81% of Class II trees (SE=0.03, N=153), and 55% of Class III trees

(SE=0.06, N=49).

Black carpenter ants were present on a significantly higher percent of trees with low numbers (0-5) of red oak borer emergence holes than trees with high numbers (5+) of red oak borer emergence holes (p<0.001, df=327) (Figure 2). Ants were present on

29% of trees with high numbers of red oak borer emergence holes (SE=0.02, N=21), whereas ants were present on 81% of trees with low numbers of red oak borer emergence holes (SE=0.09, N=307).

DISCUSSION

37 Red oak borers are less likely to be found in trees where black carpenter ants are present (Figure 2). Black carpenter ants are more abundant in apparently healthy Class

I trees than in apparently unhealthy Class III trees. There are several reasons why this trend may exist. Class I trees may provide more nutrients, moisture, or a cooler foraging environment. However, black carpenter ants may occur more frequently on Class I trees because black carpenter ants influence red oak borer population dynamics. I suggest this occurs by direct egg predation. Early literature suggested that black carpenter ants were not predatory, since they had never been observed taking live prey (Pricer 1908); however, molecular diagnostics since have confirmed that the ants are predators of red oak borer eggs, and this is not likely to be an exceptional case (Muilenberg 2008). Black carpenter ants have also been documented as predators of western spruce budworm, a notorious forest pest whose populations are capable of undergoing large-scale

explosions and crashes (Youngs and Campbell 1984).

Ito and Higashi (1991) suggest that ants may have an indirect mutualism with oak trees via myrmecophilous aphids. Ants of the genus Formica decrease the number of lepidopteran and coleopteran leaf feeders on oaks as well as the number of acorn borers in the fruit. While they amplify potentially harmful aphid populations, this still results in an overall gain in fitness for the tree because it is able to produce a higher number of viable acorns (Ito and Higashi 1991). While black carpenter ants do not domesticate any specific aphids, they do tend general populations of aphids, so a similar model of behavior may exist (Pricer 1908). Aphid tree preference could be dictating the apparent preferences of black carpenter ants that I observed.

Black carpenter ant populations seemingly would benefit from red oak borer.

Black carpenter ants nest in standing trees, and they require a point of decay to begin

38 nest excavation (Hansen and Klotz 2005). Black carpenter ants have been observed in red oak borer heartwood galleries in standing trees and down trees (Donley 1983). Red oak borers, via gallery and tunnel formation, may facilitate decay organisms, thus providing access points for black carpenter ant nest formation. However, these results do not support this hypothesis because fewer trees with previously high red oak borer populations had black carpenter ants foraging on them.

This study occurred after red oak borer populations had declined; therefore, it is difficult to draw direct conclusions about the role of black carpenter ants in the population control and/or regulation of red oak borer, and most of these data are correlative in nature. Though there is a body of literature that supports the hypothesis that black carpenter ants could be potentially important predators of forest insect pests, more direct assessments of their impact, behaviors and roles as predators is necessary before definite conclusions can be drawn.

ACKNOWLEDGEMENTS

The author thanks T. Kring, and L. Thompson for suggestions and reviews. I

also thank T. CarlLee, L. Galligan, and M. McCall for their assistance in field studies.

This project was funded in part by the Arkansas Agricultural Experiment Station, and

grants from the USDA Forest Service, Southern Research Station and USDA Forest

Service, Forest Health Protection, STDP and FHM Programs.

39 r 100 A A 90 80 B 70 60 50 40

ants present ants 30 20 10

% trees with black carpente black with % trees 0 Class I Class II Class III REP Tree Class

Figure 1: Relationship between REP tree class and black carpenter ant presence. Mean percentages for classes followed by the same letter are not significantly different (F=8.25, p=0.0003). Data shown are not statistically transformed.

40 100 A 90 80 70 60 B 50 40

ants present 30 20 10

% trees with black% carpenter 0 low high Red oak borer emergence hole class

Figure 2: Relationship between black carpenter ant presence and within-tree red oak borer populations. Means classes with the same letter are not significantly different

(F=32.82, p<0.001). Data shown are not statistically transformed.

41 Table 1: Stand characteristics for all plots surveyed, including number of red oaks in,

mean and standard error of DBH (cm) and DBH range. Summary statistics, in

bold, are presented for each geographic location and all plots.

Plot Location No. red oaks Average DBH Std Error DBH DBH Range Fly Gap 123 31.3 0.78 12.0-57.5 Fly Gap East 27 34.8 1.54 21.0-57.5 Fly Gap North 5 26.4 2.68 17.0-33.0 Fly Gap Ridge 51 29.8 1.22 12.0-51.0 Fly Gap South 40 31.3 1.37 16.5-47.0 Oark 124 36.5 1.00 11.0-73.5 Oark East 22 44.6 2.66 11.0-73.5 Oark North 28 34.3 1.89 14.5-54.0 Oark Ridge 36 32.8 1.34 11.0-51.0 Oark South 18 32.2 2.65 11.0-52.0 Oark West 20 41.0 2.28 17.0-57.0 White Rock 81 36.0 1.14 12.5-63.0 White Rock East 21 38.6 2.30 20.5-60.5 White Rock North 10 40.3 3.21 21.0-54.5 White Rock South 17 31.1 2.15 12.5-45.0 White Rock West 33 35.4 1.74 21.0-63.0

42 REFERENCES

Allen, D.C., F.B. Knight and J.L. Foltz. 1970. Invertebrate predators of the jack-pine budworm, Choristoneura pinus, in Michigan. Annals of the Entomological Society of America 63: 59-64.

Donley, D.E. and R.E. Acciavatti. 1980. Red oak borer. USDA Forest Service. Forest Insect and Disease Leaflet 163. 5 p.

Donley, D.E. 1981. Control of red oak borer by removal of infested trees. Journal of Forestry 79: 731-733.

Donley, D.E. 1983. Cultural control of the red oak borer (Coleoptera: Cerambycidae) in forest management units. Journal of Economic Entomology 76: 927-929.

Feicht, D.L. and R. Acciavatti. 1985. Pilot test of red oak borer silvicultural control in commercial forest stands. In Fifth Central Hardwood Forest Conference. Urbana, IL. 279-284.

Fierke, M.K., D.L. Kinney, V.B. Salisbury, D.J. Crook and F.M. Stephen. 2005a. Development and comparison of intensive and extensive sampling methods and preliminary within-tree population estimates of red oak borer (Coleoptera: Cerambycidae) in the Ozark Mountains of Arkansas. Environmental Entomology 34: 185-192.

Fierke, M.K., D.L. Kinney, V.B. Salisbury, D.J. Crook and F.M. Stephen. 2005b. A rapid estimation procedure for within-tree populations of red oak borer (Coleoptera: Cerambycidae). Forest Ecology and Management 215: 163-168.

Galford, J.R. 1974. Some physiological effects of temperature on artificially reared red oak borers. Journal of Economic Entomology 67: 709-710.

Galford, J.R. 1985. Role of predators on an artificially planted red oak borer population. USDA Forest Service Northeastern Forest Experiment Station. Research Note NE-331. 2 p.

Hansen, L.D. and J.H. Klotz. 2005. Carpenter ants of the United States and Canada. Ithaca, NY. Cornell University Press. 224 p.

Hay, C.J. 1969. The life history of a red oak borer and its behavior in red, black, and scarlet oak. Proceedings of the North Central Branch of the Entomological Society of America 24: 125-127.

43 Hay, C.J. 1972. Woodpecker predation on red oak borer in black, scarlet, and northern red oak. Annals of the Entomological Society of America 65: 1421-1423.

Hay, C.J. 1974. Survival and mortality of red oak borer larvae on black, scarlet, and northern red oak in eastern Kentucky. Annals of the Entomological Society of America 67: 981-986.

Heitzman, E. 2003. Effects of oak decline on species composition in a northern Arkansas forest. Southern Journal of Applied Forestry 27: 264-268.

Ito, F. and S. Higashi. 1991. An indirect mutualism between oaks and wood ants via aphids. Journal of Animal Ecology 60: 463-470.

Kilpelainen, J., P. Punttila, L. Sundstrom, P. Niemela and L. Finer. 2005. Forest stand structure, site type and distribution of ant mounds in boreal forests in Finland in the 1950s. Annales Zoologici Fennici 42: 243-258.

Muilenberg, V.L., F.L. Goggin, S.L. Hebert, L. Jia and F.M. Stephen. 2008. Ant predation on red oak borer confirmed by field observation and molecular gut- content analysis. Agricultural and Forest Entomology 10: 1-9.

Nuss, A.B., D.R. Suiter and G.W. Bennett. 2005. Continuous monitoring of the black carpenter ant, Camponotus pennsylvanicus (Hymenoptera: Formicidae), trail behavior. Sociobiology 45: 597-618.

Parmelee, F.T. 1941. Longhorned and flatheaded borers attacking fire-killed coniferous timber in Michigan. Journal of Economic Entomology 34: 377-380.

Pricer, J.L. 1908. The life history of the carpenter ant. Biological Bulletin 15: 177-218.

Riggins, J.J. 2008. Remote sensing of forest decline and Enaphalodes rufulus outbreak in the Arkansas Ozarks, U.S.A. PhD Dissertation. University of Arkansas. Fayetteville, Arkansas.

SAS. 2007. JMP Software: Version 7.0. SAS Institute, Cary, NC.

Smirnoff, W.A. 1959. Predators of Neodiprion swainei Midd. (Hymenoptera: Tenthredinidae) larval vectors of virus diseases. Canadian Entomologist 91: 246- 248.

Starkey, D., S. Mangini, F. Oliveria, S. Clarke, B. Bruce, R. Kertz and R. Menard. 2000. Forest health evaluation of oak mortality and decline on the Ozark National Forest, 1999. Forest Health Protection Report 2000-02-02. 31 p.

Stephen, F.M., V.B. Salisbury and F.L. Oliveria. 2001. Red oak borer, Enaphalodes rufulus (Coleoptera: Cerambycidae), in the Ozark Mountains of Arkansas,

44 U.S.A.: An unexpected and remarkable forest disturbance. Integrated Pest Management Reviews 6: 247-252.

USDA Forest Service. 1993. The database component of the silvicultural examination and prescription process as documented in the silvicultural examination and prescription handbook. FSH 2409.26d, R8 Amendment 2409.26d-93-1. Appendix B. Continuous Inventory of Stand Conditions, CISC. 4 p.

Ware, V.L. and F.M. Stephen. 2006. Facultative intraguild predation of red oak borer larvae (Coleoptera: Cerambycidae). Environmental Entomology 35: 443-447.

CHAPTER 3

OCCURRENCE OF BLACK CARPENTER ANTS IN OAK-HICKORY FORESTS IN RELATION TO PRESCRIBED FIRE AND SITE-STAND VARIABLES

46 INTRODUCTION

Fire is an ecosystem disturbance process, resulting naturally from lightning

strikes and other ignition sources during periods of high fuel flammability. Several

southeastern United States ecosystems including longleaf pine, oak-hickory savannas,

shortleaf pine, and tall grass prairie evolved in conjunction with fire (Brockway and

Lewis 1997, Batek et al. 1999). Of all the disturbance factors that influenced the

development of hardwood forests, fire was perhaps the most important; fires played a

dominant role in sustaining the oak-hickory ecosystem of the Ozark Highlands

(VanLear 2004).

Widespread attempts at fire exclusion began in the early twentieth century

(VanLear 2004). The first USDA Forest Service conference, the Mather Field Conference

of 1921, dealt almost exclusively with fire suppression (Jurney et al. 2004). Pre-twentieth

century fires were light-to-moderate-intensity surface fires. Increased time between fires

and accumulation of fuels since fire suppression policies were enacted has dramatically

increased the hazard for intense destructive wildfire in many stands (Guyette and

Spetich 2003, Dey et al. 2004, VanLear 2004). Fire exclusion has also caused considerable

shifts in the composition and dynamics of the Ozark oak-hickory forests, including

higher densities of trees per acre, invasion of oak stands by shade-tolerant species,

decreased oak regeneration, increased competition from fire-intolerant species and

accumulation of fuels (Brose et al. 1999, Foti 2004, VanLear 2004).

However, since 1990, prescribed fire management of oak-hickory forests has

become increasingly prevalent. In 2006, 21,853 ha of oak-hickory forests in the Ozark

National Forest of Arkansas were treated with prescribed fire. This is a significant

47 increase from just two years prior, when only 12,140 ha were managed with prescribed fire, though this is still a small fraction of the 485,622 ha that compose the Ozark

National Forest (Andre et al. 2007). Prescribed fire is a management tool that clears dead and dying plant materials, reduces understory vegetation, tree density and the invasion of exotic plants, increases species diversity, aids in the control of plant diseases and restores fire-dependent ecosystems (Ahlgren and Ahlgren 1960, Habeck and Mutch

1978, Brockway and Lewis 1997, McRae 1994, Foti 2004).

Habitat disturbances are important influences on the composition and structure of ant communities (Andrew et. al 2000). The effects of prescribed fire on ants

(Hymenoptera: Formicidae) have been studied in, among others, ponderosa pine, tropical cloud forests, desert grasslands and agroecosystems (Zimmer and Parmenter

1998, Schonberg et al. 2004, Castano-Meneses and Palacios-Vargas 2003, Stephens and

Wagner 2006). In agricultural crops, ant feeding guilds shift from nectivorous and omnivorous species to only omnivorous species after slashing and burning (Castano-

Meneses and Palacios-Vargas 2003). In pine barrens, no difference in the abundance of black carpenter ants (Camponotus pennsylvicanus DeGeer) was found between burned and unburned sites (Buffington 1967). It appears that the effects of both wildland and prescribed fire on ant species composition and abundance vary by ecosystem and the intensity of and interval between fires.

Ants are also affected by other management practices. Thinning and deforestation have been shown to cause changes in ant abundance and diversity

(Schonberg et al. 2004, Sorvari 2007). Ant communities are also influenced by habitat type though tree species composition is not thought to be an important variable affecting most ant communities, since they nest in fallen logs and leaf litter (Yanoviak and

48 Kaspari 2000, Hill et al. 2008). However, black carpenter ants nest in and forage on standing trees, so tree species may be important in predicting their abundance and diversity (Hansen and Klotz 2005).

Ants, particularly carpenter ants, have also been implicated as predators of forest pest insects. In a study in which wood-boring larvae were artificially planted in fire- killed conifers, researchers noted that data collection was particularly difficult, because

“red ants” ate the larvae before they could be checked (Parmelee 1941). When carpenter ants were excluded from trees infested with western spruce budworm pupae, 8% of the pupae were removed, whereas, when carpenter ants were not excluded, 85% of the pupae were removed (Youngs and Campbell 1984). Carpenter ants have also been observed as predators of jackpine budworm, Choristoneura pinus Freeman, and the

Swaine jackpine sawfly, Neodiprion swainei Midd. (Smirnoff 1959, Allen et al. 1970).

Black carpenter ants are known predators of the red oak borer, Enaphalodes rufulus

Haldeman (Coleoptera: Cerambycidae), a native wood-boring beetle that underwent a dramatic and unprecedented population increase and decrease in the early twenty-first century in the oak-hickory forests of the Arkansas Ozarks (Muilenberg et al. 2008). The impact of black carpenter ants as predators of the red oak borer during outbreak conditions is unknown; however, at endemic levels black carpenter ants are estimated to be responsible for 13-29% of the total larval predation (Galford 1985).

With the increasing occurrence of prescribed fire, it is important to

comprehensively evaluate its effects on insect populations, particularly those that may

be potentially significant predators of forest pests. The effects of management practices,

including prescribed fire, vary by habitat type, so it is also important to evaluate

treatments on a local level (Ratchford 2005). Since black carpenter ants are known

49 predators of red oak borers and prescribed fire is becoming more important in Ozark oak-hickory forests, the effects of fire on black carpenter ants should be evaluated in this ecosystem. The objective of this study was to evaluate the effects of dormant season prescribed fire and site and stand variables on the distribution of black carpenter ants in oak-hickory forests of the Ozark Mountains of Arkansas.

METHODS

Site Description

Nine sites were selected in the Ozark National Forest on the basis of their recent management treatments. Three sites were treated with dormant season prescribed fire in each of 2005, 2006, and 2007. Each of these sites was paired with an adjacent control site which had not been treated with fire or thinned since 2000 (Table 1). These sites were situated in oak-hickory dominated forests on multiple aspects and slope gradients.

All sites were covered by at least 50% oak (Quercus species.).

Fire Effects

All fire-treated sites were burned during the dormant season (November- March) using non-aerial lighting protocols. These fires were low to moderate intensity surface fires that were ignited to reduce fuel loading (e.g. leaf litter layer). All fires burned in

the same fuel model: oak-hickory closed canopy forest. No extreme fire effects, such as crown fire or complete stand mortality, were observed in any site. Fire-induced lesions

50 or fire-killed buds, collectively known as “scorch”, did not exceed 12 feet on any of the trees sampled, and only reached heights greater than eight feet on five of the trees sampled.

Direct Observations of Ants

These data were analyzed with a 3x3x4 three-way factorial design. Fire treatment was tested at four levels (burned in 2005, 2006, 2007 and unburned). Tree species was tested at three levels (red oak, white oak and hickory). Tree species was tested at three levels (small, medium and large). Within each burned and control site, five 30x30 meter (0.09 ha) square plots were randomly selected and established for a total of 90 plots. In each plot, all live standing trees larger than ten centimeters diameter at breast height (1.3 m above the ground) were identified and marked. The species, diameter, and fire treatment for each tree were recorded.

Trees were baited using five mL of a 1:1 mixture of tuna in oil, and clover honey.

Baits were placed in 10 mL paper cups and affixed to trees at breast height using aluminum nails. The bait was left undisturbed in the plot for one hour, then ants in and near baits on the tree were collected and returned to the laboratory for identification.

All trees were evaluated for ants on a presence/absence basis, because the number of

ants present on a tree gives little indication of the number of colonies or individuals per unit area (Kilpelainen et al. 2005). Trees were sampled during the hours of 06:00-11:00 from May through September, as black carpenter ants are active during this period and foraging is likely to be negatively affected by extreme afternoon heat (Hansen and Klotz

2005). A subset of trees (n=100) was sampled twice to insure test fidelity.

Data Analysis

The dominant tree species groups were red oaks (Section Lobatae), white oaks

(Section Quercus) and hickories (Carya species). These three groups were exclusively used to analyze tree species effects. Trees were grouped into three classes based on their diameters at breast height (DBH). Trees that were less than 20 cm DBH were classified as “small.” Trees 20 cm to 30.5 cm DBH were classified as “medium.” Trees greater than 30.5 cm DBH were classified as “large.” Trees were also grouped by fire treatment into areas burned in 2005, 2006 and 2007, and unburned (control) areas. Data were summarized as percentages of trees of a given attribute (e.g. size class, species, fire treatment) with black carpenter ants present. These percentages were normalized using an arc-sine transformation. Interactions among fire treatment, tree species, and tree diameter classes were evaluated using a standard least squares model and one-way

ANOVA (α=0.05). Variables that had significant interactions were further analyzed using Tukey pair-wise comparison tests. Results were analyzed using JMP 7.0 (SAS

2007). Voucher specimens were deposited in the Insect Museum at the University of

Arkansas, Fayetteville, Arkansas.

RESULTS

Ants were surveyed on a total of 3556 trees. Of these 25% were red oaks (n=881),

35% were white oaks (n=1255), 20% were hickory (n=709), and 20% were other species,

52 primarily of the genera Acer, Cornus, and Fraxinus (n=711). The species composition of areas burned in 2005, 2006, 2007, and unburned areas were not significantly different

(p=0.06, df=2). The average DBH for these trees was 23.5 cm; half of all trees were 17-28 cm DBH, and DBH ranged from 10-80 cm.

Average stem density over all plots was 439 trees per ha, with a range of 322 to

678 trees per ha (29-61 trees per 30x30 m plot). Areas burned in 2005, 2006, 2007, and areas that were not burned did not have significantly different numbers of trees per ha

(p=0.12, df=3).

Of the 100 trees that were re-surveyed as a verification test for ant fidelity, 95 showed the same result during both sampling trials. Over all plots, the mean percentage of trees with black carpenter ants present on them was 55 (SE=0.17).

Interactions Among Tree Species, Tree Diameter Class, Prescribed Fire and Black

Carpenter Ant Distribution

There was no interaction between fire treatment and tree species or tree diameter class (p=0.05, df=6; p=0.57, df=6, respectively).

The combination of tree species and tree diameter class significantly affected black carpenter ant distribution (p<0.001, df=6) (Table 2). Ant distribution was significantly different among trees of different species within each diameter class.

53 Black Carpenter Ant Abundance by Fire Treatment

Black carpenter ants were present on significantly different percentages of trees in areas burned in 2005, 2006 and 2007 (p<0.001, df=89) (Figure 1). The percent of trees with black carpenter ants present in unburned controls and areas burned in 2007 did not significantly differ. Ants were present on a mean 52% of trees in areas burned in 2005

(SE=0.02, N=575), on a mean 28% of trees in areas burned in 2006 (SE=0.02, N=448) and on a mean 65% of trees in areas burned in 2007 (SE=0.02, N=418). Ants were present on a mean 62% of trees in unburned control areas (SE=0.01, N=1404).

DISCUSSION

Black carpenter ant abundance differed on trees of different species, different size classes, and different fire regimes. Black carpenter ants appear to prefer large trees and oaks and are least commonly found on hickories and small trees. In this study, large red oaks are most likely to have black carpenter ants present, and small hickories are least likely. Foliar chemistry, such as levels of tannins, foliar carbohydrates, fiber, and nutrients vary by tree species (Rieske et al. 2002). A substantial portion of the diet of black carpenter ants comes from aphid honeydew, which is extracted from foliar phloem (Pricer 1908). Variability in nutrient quality among tree species may influence aphid density and diversity and thus influence ants’ tree preferences; however, we do not know of any work that supports or denies this hypothesis, or of comparisons of foliar chemistry among multiple species of oaks, between oaks and hickories, and within

54 tree size and/or age classes. To further complicate the matter, foliar chemistry is highly variable and changes are inducible via feeding, disease, fire and weather (Rieske et al.

2002, Holton et al. 2003, Rieske et al. 2003).

Time elapsed since a prescribed burn significantly correlated to presence of black carpenter ants in oak-hickory forests (Figure 1). Results were unpredictable: Areas that were burned 6 months before evaluation and controls were not different. Black carpenter ants were found on significantly fewer trees in areas burned 1.5 years before evaluation than control areas. Areas burned 2.5 years before evaluation had fewer trees with black carpenter ants than control areas, but greater numbers of trees with black carpenter ants than areas burned 1.5 years before evaluation.

Since black carpenter ant presence was not affected immediately after a fire and did not show lower numbers until 1.5 years post burn, it would appear that habitat modifications are a likely cause of these differences. Identifying potential habitat

modifications that caused changes in black carpenter ant abundance is beyond the scope

of this research; however, ant abundance could have decreased due to increased

competition from fire-tolerant ant species, increased predation by woodpeckers

(Picidae), shortages in resources such as honeydew-producing aphids, changes in nest

site suitability by removal of rotting trees and/or other unknown biotic or abiotic alterations that resulted from fire.

Previous studies have had contrasting results relative to the effects of wildfire, thinning, prescribed fire, and prescribed fire/thinning combinations on the abundance and diversity of ants with contrasting results (Hanula and Wade 2003, Ratchford et al.

2005, Stephens and Wagner 2006). Wood ants (Formica spp.) decreased in abundance

after deforestation via commercial logging (Sorvari and Hakkarainen 2007). Ant genera

55 richness and abundance were unaffected immediately after, but decreased in the year following a fire in Sky Island communities (Wilkinson et al. 2005). Ant species assemblages shifted after wildfires, thinning and prescribed fire in Arizona ponderosa pine (Stephens and Wagner 2006). Carpenter ants (Camponotus spp.) were affected by fire intervals in Georgia longleaf pine savannas, with the lowest numbers occurring in areas that had been burned annually (Hanula and Wade 2003). These studies also showed that dormant season fire decreased the number of insect predators in a community (Hanula and Wade 2003). In the upland forests of northern California, the time since and intensity of a disturbance were important in predicting ant species richness, though the reasons for this were unclear (Ratchford et al. 2005).

An overriding theme of all these studies is that changes in ant communities in

burned areas appear to be manifested through fire-induced habitat alterations and not

direct flame mortality. The location in which the study occurred, the species studied,

the nature and season of the fire and previous management of the study area all affect

the magnitude and nature of these habitat alterations; therefore, it is important to

evaluate these effects on a local basis (Hanula and Wade 2003). To my knowledge, this

study is the first to evaluate black carpenter ant presence in upland oak-hickory forests

and one of few fire-insect interaction studies in oak-hickory forests.

All fires utilized in this study occurred in dormant seasons; however, effects of

fire on insects have been shown to vary by season (New and Hanula 1998). Future

studies could investigate the long-term effects of fire on insect populations and the

relationship between burn season and insect abundance and diversity. Ants have strong

potential as an indicator species, but research in this area is sparse and warrants further

investigation (Stephens and Wagner 2006). This work demonstrates that ants are

56 sensitive to fire disturbances in oak hickory forests; however, further work should also

be directed at understanding the ways in which habitat alterations are influencing insect

populations.

ACKNOWLEDGMENTS

The author thanks T. Kring and L. Thompson for suggestions and reviews. I also

thank T. CarlLee, L. Galligan, and M. McCall for their assistance in field studies. This

project was funded in part by the Arkansas Agricultural Experiment Station, and grants

from the USDA Forest Service, Southern Research Station and USDA Forest Service,

Forest Health Protection, STDP and FHM Programs.

90 C C

70 A

50 B 40 present 30 20

% trees with black carpenter ants ants carpenter black with trees % 0

2005 2006 2007 Control Year burned

Figure 1: Effects of Fire Treatment on Black Carpenter Ant Abundance. Fire treatment affects the percent of trees with black carpenter ants present (F=111.61, p<0.001). Bars connected by a letter are not significantly different. Data shown have not been statistically transformed.

58 Table 1. Geographic locations, fire treatments, and stand characteristics for all areas used in this study. Shown are geographic coordinates, fire treatments, tree species composition, and tree numbers for each area surveyed. All UTM coordinates are in

UTM zone 15-N NAD83.

Plot Code N (trees Burn Year Avg % % % % UTM in plot) DBH Red White Hickory Other Coordinates (cm) Oak Oak Species BURN2005-1 246 2005 24.3 19.9 60.6 11.4 8.1 03968252, 451358 BURN2005-2 163 2005 23.8 14.7 41.7 42.9 <1.0 03967487, 449922 BURN2005-3 199 2005 24.2 20.1 48.2 25.6 6.0 03986667, 452654 BURN2006-1 144 2006 26.0 10.4 29.2 56.3 3.5 03964408, 453497 BURN2006-2 176 2006 28.1 21.6 39.2 30.7 8.5 03958066, 483174 BURN2006-3 156 2006 26.1 19.2 34.6 41.7 4.5 03964759, 443510 BURN2007-1 222 2007 21.1 28.4 27.5 10.4 33.8 03958618, 494988 BURN2007-2 209 2007 24.9 33.5 26.8 14.4 25.4 03934229, 492238 BURN2007-3 172 2007 32.2 12.8 41.9 12.2 33.1 03948067, 478223 CONTROL-1 213 NA 24.6 23.9 33.8 24.4 17.8 03965063, 443113 CONTROL-2 186 NA 25.2 30.7 31.2 11.8 26.3 03961596, 481906 CONTROL-3 162 NA 24.8 26.5 33.3 19.1 21.0 03962715, 453667 CONTROL-4 214 NA 24.7 29.4 22.4 10.3 37.9 03948113, 478384 CONTROL-5 186 NA 24.9 22.0 33.3 18.8 25.8 03968229, 451114 CONTROL-6 240 NA 25.6 38.8 26.7 9.6 25.0 03967896, 449974 CONTROL-7 242 NA 24.2 25.6 30.2 12.8 31.4 03934285, 492177 CONTROL-8 205 NA 24.6 23.9 33.7 12.2 30.2 03958318, 495495 CONTROL-9 260 NA 24.8 27.3 33.9 17.3 21.5 03975349, 449059

59 Table 2. Effect of tree species and size class on black carpenter ant presence. The interaction among tree species, tree diameter class and black carpenter ants is significant

(F=13.04, p<0.001). Means followed by the same letter are not significantly different.

Data shown are not statistically transformed.

Species/Size Class Small Medium Large Red oak 72% D 92% B 94% A

(SE=0.04, N=133) (SE=0.02, N=555) (SE=0.03, N=193)

White oak 32% D 67% C 83% A

(SE=0.02, N=312) (SE=0.02, N=713) (SE=0.03, N=230)

Hickory 25% E 35% D 40% B,C

(SE=0.03, N=267) (SE=0.02, N=350) (SE=0.04, N=92

60 REFERENCES

Ahlgen, I.F. and C.E. Ahlgren. 1960. Ecological effects of forest fires. Botanical Review 483-517.

Allen, D.C., F.B. Knight and J.L. Foltz. 1970. Invertebrate predators of the jack-pine budworm, Choristoneura pinus, in Michigan. Annals of the Entomological Society of America 63: 59-64.

Andre, John, M. Morales and M. Anderson. 2007. Restoring the Interior Ozark Highlands. Fire Management Today 67: 20-23.

Andrew, N., L. Rodgerson and A. York. 2000. Frequent fuel-reduction burning: the role of logs and associated leaf litter in the conservation of ant biodiversity. Australian Ecology 25: 99-107.

Brose, P., D. VanLear and R. Cooper. 1999. Using shelterwood harvests and prescribed fire to regenerate oak stands on productive upland sites. Forest Ecology and Management 113: 125-141.

Buffington, J.D. 1967. Soil arthropod populations of the New Jersey pine barrens as affected by fire. Annals of the Entomological Society of America 60: 530-535.

Campbell, J.W., J.L. Hanula and T.A. Waldrop. 2007. Effects of prescribed fire and fire surrogates on floral visiting insects of the Blue Ridge province in North Carolina. Biological Conservation 134: 393-404.

Castano-Meneses, G. and J.G. Palacios-Vargas. 2003. Effects of fire and agricultural practices on neotropical ant communities. Biodiversity and Conservation 12: 1913-1919.

Dey, D.C., R.P. Guyette and M.C. Stambaugh. 2004. Fire history of a forest, savanna, and fen mosaic at White Ranch State Forest. In: Spetich, Martin A., ed. 2004. Upland oak ecology symposium: history, current conditions, and sustainability. Gen. Tech. Rep. SRS-73. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 132-137.

Foti, T.L. 2004. Upland hardwood forests and related communities of the Arkansas Ozarks in the early 19th century. In: Spetich, Martin A., ed. 2004. Upland oak

61 ecology symposium: history, current conditions, and sustainability. Gen. Tech. Rep. SRS-73. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 21-29.

Galford, J.R. 1985. Role of Predators on an Artificially Planted Red Oak Borer Population. USDA Forest Service Northeastern Forest Experiment Station. Research Note NE-331. 2 p.

Guyette, R.P. and M.A. Spetich. 2003. Fire history of oak-pine forests in the lower Boston Mountains, Arkansas, USA. Forest Ecology and Management 180: 463- 474.

Habeck, J.R. and R.W. Mutch. 1973. Fire-dependent forests in the northern Rocky Mountains. Quaternary Research 3: 408-424.

Hansen, L.D. and J.H. Klotz. 2005. Carpenter ants of the United States and Canada. Ithaca, NY. Cornell University Press. 224 p.

Holton, M.K., R.L. Lindroth and E.V. Nordheim. 2003. Foliar quality influences tree-herbivore-parasitoid interactions: effects of elevated CO2 and O3 and plant genotype. Oecologia 137: 233-244.

Hill, J.G., K.S. Summerville and R.L. Brown. 2008. Habitat associations of ant species (Hymenoptera: Formicidae) in a heterogeneous Mississippi landscape. Environmental Entomology 37: 453-463.

McRae, D.J. 1994. Prescribed fire converts spruce budworm-damaged forest. Journal of Forestry 92: 38-40.

Mutch, R.W. 1970. Wildland fires and ecosystems—a hypothesis. Ecology 51: 1046- 1051.

62 New, K.C. and J.L. Hanula. 1998. Effect of time elapsed after prescribed burning in longleaf pine stands on potential prey of the red-cockaded woodpecker. Southern Journal of Applied Forestry 22: 175-183.

Parmelee, F.T. 1941. Longhorned and flatheaded borers attacking fire-killed coniferous timber in Michigan. Journal of Economic Entomology 34: 377-380.

Pricer, J.L. 1908. The Life History of the Carpenter Ant. Biological Bulletin 15: 177- 218.

Ratchford, J.S., S.E. Wittman, E.S. Jules, A.M. Eillison, N.J. Gotelli and N.J. Sanders.2005. The effects of fire, local environment and time on ant assemblages in fens and forests. Diversity and Distributions 11: 487-497.

Rieske, L.K., H.H. Housman and M.A. Arthur. 2002. Effects of prescribed fire on canopy foliar chemistry and suitability for an insect herbivore. Forest Ecology and Management 160: 177-187.

Rieske, L.K., C.C. Rhoades and S.P. Miller. 2003. Foliar chemistry and gypsy moth, Lymantria dispar (L.), herbivory on pure American chestnut, Castanea dentate (Fam: Fagacaeae) and a disease-resistant hybrid. Environmental Entomology 32: 359-365.

SAS. 2007. JMP Software: Version 7.0. SAS Institute, Cary, NC.

Schonberg, L.A., J.T. Longino, N.M. Nadkarni, S.P. Yanoviak and J.C. Gering. 2004. Arboreal ant species richness in primary forest, secondary forest, and pasture habitats of a tropical montane landscape. Biotropica 36: 402-409.

Smirnoff, W.A. 1959. Predators of Neodiprion swainei Midd. (Hymenoptera: Tenthredinidae) larval vectors of virus diseases. Canadian Entomologist 91: 246-248.

Sorvari, J. and H. Hakkarainen. 2007. Wood ants are wood ants: deforestation causes population declines in the polydomous wood ant Formica aquilonia. Ecological Entomology 32: 707-711.

VanLear, D.H. 2004. Upland oak ecology and management. In Upland Oak Ecology Symposium. Gen. Tech. Rep. SRS-73. Asheville, NC: USDA, M.A. Spetich ed. Forest Service Southern Research Station. 65-71

63 Wilkinson, E.B., E.G. Lebrum, M.L. Spencer, C. Whitby and C. Kleine. 2005. Short- term effects of fire on Sky Island ant communities. In USDA Forest Service Proceedings RMRS-P-36. Boulder, CO. 550-552.

Yanoviak, S.P. and M. Kaspari. 2000. Community structure and the habitat template: ants in the tropical forest canopy and litter. Oikos 89: 259-266.

Zimmer, K. and R.R. Parmenter. 1998. Harvester ants and fire in a desert grassland: Ecological responses of Pogonomyrmex rugosus (Hymenoptera: Formicidae) to experimental wildfires in central New Mexico. Environmental Entomology 27: 282-287.

CHAPTER 4

SUMMARY AND FUTURE DIRECTIONS

65 Summary

My objectives were to examine red oak borer emergence holes and site aspects and correlate these forest and tree attributes with presence or absence of black carpenter ants.

Historic red oak borer population data, tree REP class and site aspects for 13 separate plots were used. At each site, all red oaks >10 cm DBH were baited for black carpenters ants using a mixture of tuna in oil and honey. Black carpenter ants are more likely to be found on trees with low levels of previous red oak borer infestation versus those trees with previously high levels of infestation. These results may suggest black carpenter ants play a role in controlling red oak borer populations. Future investigations should be directed at efforts to understand whether black carpenter ants simply prefer different apparently healthy trees or if, via predation, these ants are acting as agents of red oak borer control.

I also examined how tree species and size, site and stand variables, and management practices influence black carpenter ant abundance. Fire treatment, tree species, and tree size were described for 18 plots. All trees were baited as described above, and black carpenter ant presence/absence was recorded for each tree. Black carpenter ants were more commonly present on oaks than on hickories and appear to prefer large trees over small trees. Time elapsed since a prescribed burn appears to be important in determining black carpenter ant presence, potentially via fire-induced habitat modifications, although further investigation is necessary to either confirm or refute this hypothesis.

66 Future Directions

Several projects can be developed from the research reported herein. A variety of unanswered questions remain, and several preliminary conclusions suggested in our discussions can be further explored in hopes of finding more definitive explanations for the observed phenomena. Unanswered questions remain about the natural history and ecological role of black carpenter ants in forests. There is little known about their potential to promote decay and their potential as agents of mortality to other forest insects. My research also suggested that the healthiest, largest, trees are most likely to have black carpenter ants present.

Further exploration to discover if ants may play a role in promoting tree health would be valuable.

My research suggests that disturbances such as prescribed fire influence black carpenter ant occurrence, though the possibility of ants as indicators of ecosystem health in North

American forests is relatively unexplored, and I know of no research that has examined carpenter ants as indicators of ecosystem health. Ants, as a group, have been explored as indicators of ecosystem health; however, this research is limited mostly to Australia and the

South American tropics.

I demonstrated that black carpenter ants exhibit tree size and species preferences, but the reasons behind these trends are still unapparent. I suggested that ant tree preference may in some way be dictated by aphid tree preference; however, there is no research to substantiate or refute this hypothesis; future studies could examine these relationships. While I suggest that habitat alterations are the way in which prescribed fire influences occurrence of black carpenter ants, the mechanism or type of habitat alteration is yet unknown, and future work could be directed at exploring how and why fires affect ant occurrence. Finally, fire-insect interactions

67 are still a vastly understudied field. There are only limited and highly variable reports that relate the effects of wildland or prescribed fire on most arthropod populations. As prescribed fire becomes more prevalent on the landscape, these effects become increasingly important and should be studied in greater depth.