RELATIONSHIPS BETWEEN DEAD WOOD AND ARTHROPODS IN THE

SOUTHEASTERN UNITED STATES

by

MICHAEL DARRAGH ULYSHEN

(Under the Direction of James L. Hanula)

ABSTRACT

The importance of dead wood to maintaining forest diversity is now widely recognized.

However, the habitat associations and sensitivities of many species associated with dead wood remain unknown, making it difficult to develop conservation plans for managed forests. The purpose of this research, conducted on the upper coastal plain of South Carolina, was to better understand the relationships between dead wood and arthropods in the southeastern United

States. In a comparison of forest types, more beetle species emerged from logs collected in upland pine-dominated stands than in bottomland hardwood forests. This difference was most pronounced for Quercus nigra L., a species of tree uncommon in upland forests. In a comparison of wood postures, more beetle species emerged from logs than from snags, but a number of species appear to be dependent on snags including several canopy specialists. In a study of saproxylic beetle succession, species richness peaked within the first year of death and declined steadily thereafter. However, a number of species appear to be dependent on highly decayed logs, underscoring the importance of protecting wood at all stages of decay. In a study comparing litter-dwelling arthropod abundance at different distances from dead wood, arthropods were more abundant near dead wood than away from it. In another study, ground- dwelling arthropods and saproxylic beetles were little affected by large-scale manipulations of dead wood in upland pine-dominated forests, possibly due to the suitability of the forests surrounding the plots.

INDEX WORDS: dead wood, coarse woody debris, saproxylic, Coleoptera, litter-dwelling arthropods, ground-dwelling arthropods, snags, logs RELATIONSHIPS BETWEEN DEAD WOOD AND ARTHROPODS IN THE

SOUTHEASTERN UNITED STATES

by

MICHAEL DARRAGH ULYSHEN

B.S., Miami University, 2002

M.S., University of Georgia, 2005

A Dissertation Submitted to the Graduate Faculty of The University of Georgia in Partial

Fulfillment of the Requirements for the Degree

DOCTOR OF PHILOSOPHY

ATHENS, GEORGIA

2009 © 2009

Michael Darragh Ulyshen

All Rights Reserved RELATIONSHIPS BETWEEN DEAD WOOD AND ARTHROPODS IN THE

SOUTHEASTERN UNITED STATES

by

MICHAEL DARRAGH ULYSHEN

Major Professor: James L. Hanula

Committee: Joseph McHugh Darold Batzer

Electronic Version Approved:

Maureen Grasso Dean of the Graduate School The University of Georgia May 2009 ACKNOWLEDGEMENTS

I thank my advisor Jim Hanula and committee members Joe McHugh and Darold Batzer for overseeing this research. The following people were of invaluable assistance in the field or lab: Scott Horn, Danny Dyer, Mike Cody, Stephanie Cahill, Margo Briesch, Walter Sikora, Ryan

Malloy, Jared Swain and Mary Williams. I also wish to thank Harry Lee, Bob Rabaglia, Alexey

Tishechkin, and Chris Carlton for assistance in identifying Carabidae, Scolytinae, Histeridae, and

Pselaphinae, respectively. I am grateful to Cecil Smith and Mike Thomas for their assistance and hospitality during visits to the Georgia Museum of Natural History and the Florida State

Collection of Arthropods, respectively. Finally, I thank and am in particular debt to the Ulyshens and Keisters. Support for this research was provided by the Department of Energy-Savannah

River Operations Office through the U.S. Forest Service Savannah River under Interagency

Agreement DE-AI09-00SR22188.

iv TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS...... iv

CHAPTER

1 INTRODUCTION AND LITERATURE REVIEW...... 1

2 HABITAT ASSOCIATIONS OF SAPROXYLIC BEETLES IN THE

SOUTHEASTERN UNITED STATES: A COMPARISON OF FOREST TYPES,

TREE SPECIES AND WOOD POSTURES...... 3

3 RESPONSES OF ARTHROPODS TO LARGE SCALE MANIPULATIONS OF

DEAD WOOD IN LOBLOLLY PINE STANDS OF THE SOUTHEASTERN

UNITED STATES...... 47

4 PATTERNS OF SAPROXYLIC BEETLE SUCCESSION IN LOBLOLLY PINE..83

5 CONCLUSIONS...... 117

v CHAPTER 1

INTRODUCTION AND LITERATURE REVIEW

Dead wood is an important resource and structural element in terrestrial, aquatic, estuarine, coastal beach, and open ocean habitats throughout the world (Maser et al. 1988). In terrestrial systems dead wood plays an important role in nutrient cycling and provides habitat for many plants, animals, fungi, and bacteria (Harmon et al. 1986). Terrestrial animal communities found within dead wood are dominated by arthropods, with beetles being the most diverse of all.

It has been estimated that about 20-25% of all beetle species are saproxylic, meaning they cannot persist without dead or dying wood (Elton 1966, Grove 2002). Because timber harvesting operations inevitably reduce the volume and diversity of dead wood, saproxylic beetles have been severely impacted by forest management in many parts of the world. In Finland, for example, the amount of dead wood across the landscape has been reduced by 90-98%, putting over half of all saproxylic species at risk of disappearance (Siitonen 2001). Unfortunately, biodiversity research in North America is in its infancy and the status of most arthropods remains completely unknown. However, we do have some idea how many beetle species are present and potentially at risk at both continental and regional scales. North America north of Mexico contains approximately 25,160 beetle species (Marske and Ivie 2003) and the relatively small state of South Carolina, where this research took place, contains at least 3,346 species (Peck and

Thomas 1998, and references therein). Assuming that 20-25% of all beetle species depend on dead wood (Elton 1966, Grove 2002), it can be estimated that South Carolina alone contains between 669 and 836 saproxylic beetle species! Research is urgently needed to better understand

1 the status and requirements of this diverse community if the biological catastrophe experienced in the boreal forests of northern Europe is to be avoided in the southeastern United States.

Literature Cited

Elton, C. S. 1966. The Pattern of Animal Communities. Methuen and Co Ltd, London.

432 p.

Grove, S. J. 2002. Saproxylic insect ecology and the sustainable management of forests. Annual

Review of Ecology and Systematics. 33: 1-23.

Harmon, M. E., Franklin, J. F., Swanson, F. J., Sollins, P., Gregory, S. V., Lattin, J. D.,

Anderson, N. H., Cline, S. P., Aumen, N. G., Sedell, J. R., Lienkaemper, G. W.,

Cromack, K., Jr., Cummins, K. W. 1986. Ecology of coarse woody debris in temperate

ecosystems. Advances in Ecological Research 15: 133-302.

Marske, K. A., Ivie, M. A. 2003. Beetle fauna of the United States and Canada. The

Coleopterists Bulletin 57: 495-503.

Maser, C., Tarrant, R. F., Trappe, J. M., Franklin, J. F. (tech. eds.). 1988. From the forest to the

sea: a story of fallen trees. USDA Forest Service General Technical Report PNW-GTR-

229. 153 p.

Peck, S. B., Thomas, M. C. 1998. A distributional checklist of the beetles (Coleoptera) of

Florida. Arthropods of Florida and neighboring land areas. Vol. 16. 180 pp.

Siitonen, J. 2001. Forest management, coarse woody debris and saproxylic organisms:

Fennoscandian boreal forests as an example. Ecological Bulletins 49: 11-41.

2 CHAPTER 2

HABITAT ASSOCIATIONS OF SAPROXYLIC BEETLES IN THE SOUTHEASTERN

UNITED STATES: A COMPARISON OF FOREST TYPES, TREE SPECIES AND WOOD

POSTURES¹

¹Ulyshen, M.D. and J.L. Hanula. 2009. Forest Ecology and Management 257: 653-664.

Reprinted here with permission of publisher.

3 Abstract

Saproxylic beetles are highly sensitive to forest management practices that reduce the abundance and variety of dead wood. Yet this diverse fauna continues to receive little attention in the southeastern United States even though this region supports some of the most diverse, productive and intensively managed forests in North America. In this replicated three-way factorial experiment, we investigated the habitat associations of saproxylic beetles on the coastal plain of South Carolina. The factors of interest were forest type (upland pine-dominated vs. bottomland hardwood), tree species (Quercus nigra L., Pinus taeda L. and Liquidambar styraciflua L.) and wood posture (standing and downed dead wood, i.e., snags and logs). Wood samples were taken at four positions along each log and snag (lower bole, middle bole, upper bole and crown) ~11 months after the trees were killed and placed in rearing bags to collect emerging beetles. Overall, 33,457 specimens from 51 families and ≥ 250 species emerged.

Based on an analysis of covariance, with surface area and bark coverage as covariates, saproxylic beetle species richness differed significantly for forest type and wood posture. There were no significant interactions. Species richness was significantly higher in the upland pine-dominated stand than the bottomland hardwood forest, possibly due to higher light exposure in upland forests. Although L. styraciflua yielded more beetle species (152) than either Q. nigra (122) or

P. taeda (125), there were no significant differences in species richness among tree species.

There were also no relationships evident between relative tree abundance and observed or expected beetle species richness. Significantly more beetle species emerged from logs than from snags. However snags had a distinct fauna including several potential canopy specialists. Our results suggest that conservation practices that retain or create entire snags as opposed to tall stumps or logs alone will most greatly benefit saproxylic beetles in southeastern forests.

4 Keywords: biodiversity, coarse woody debris, dead wood, forest canopy, resource partitioning, vertical stratification

Introduction

Although it is now widely recognized that saproxylic beetles are highly sensitive to long- term losses of dead wood, virtually nothing is known about the status of this diverse community in the intensively managed forests of the southeastern United States. It seems doubtful, however, that a group so vulnerable to human activities could be doing well in a country in which one- third of all plant and animal species are thought to be at risk (Stein et al., 2000). Furthermore, the southeastern region faces a number of continuing (e.g., timber harvesting), intensifying (e.g., urbanization and habitat fragmentation), and emerging (e.g. biofuel production) threats to saproxylic organisms. These ever-younger and smaller forests may require new management strategies if we wish to retain a diverse community of saproxylic beetles as well as other dead wood inhabiting organisms. Unfortunately, too little is known about the basic life histories and habitat requirements of most species to prioritize actions or to make informed decisions. Here we investigate the habitat associations of saproxylic beetles on the upper coastal plain of the southeastern United States. The main factors of interest are summarized below.

1. Forest type. The coastal plain of the southeastern United States is dominated by pines on relatively dry upland sites and by mixed hardwoods on mesic bottomland sites. The relative importance of these two main forest types to saproxylic beetles remains unknown. Upland pine forests are more extensive than bottomland hardwood forests throughout the region. However, bottomland hardwood forests support more diverse tree species assemblages and may therefore be disproportionately important to the saproxylic beetle fauna. A number of factors likely to

5 differ between forest types, such as canopy coverage (Økland, 2002), light exposure (Lindhe et al., 2005) and humidity (Warriner et al., 2004), may also have important consequences for the structure and species richness of saproxylic beetle communities. In this study we sampled saproxylic beetles in both an upland and bottomland forest. We predicted that overall beetle richness would be higher in the bottomland forest than in the upland forest due to the higher diversity of tree species in bottomland forests.

2. Tree species. We sampled wood from three tree species, each of which differed in abundance between upland and bottomland forests, to evaluate the effects of relative tree abundance on the diversity and composition of saproxylic beetles. This question has particularly important implications for saproxylic beetle conservation, but remains largely unstudied. We predicted a significant interaction between tree species and forest type due to differences in relative tree species abundances between the two forest types.

3. Wood posture. A large volume and variety of resources are available to saproxylic insects above the ground in the form of standing dead trees (i.e., snags), dead branches and twigs, and rotting heart wood (Fonte and Schowalter, 2004). For example, in a temperate broadleaved forest in Sweden, Nordén et al. (2004) found snags made up about 22% of total dead wood volume and another 6% was attributed to dead branches attached to living trees. Standing or suspended dead wood is generally drier and decays more slowly than wood in contact with the ground (Jomura et al., 2008), possibly reducing the abundance and diversity of insects present

(Larkin and Elbourn, 1964). Several studies from Europe support this notion (Jonsell and

Weslien, 2003; Gibb et al., 2006; McGeoch et al., 2007; Hjältén et al., 2007; Franc, 2007).

However, many threatened species and other insects appear to favor snags (Jonsell et al.,1998;

Sverdrup-Thygeson and Ims, 2002; Kappes and Topp, 2004; Hedgren and Schroeder, 2004).

6 Unfortunately, previous efforts to sample from snags have generally limited sampling to within a few meters of the ground. Until the upper reaches of snags are adequately sampled, it will be impossible to reach definite conclusions regarding the relative importance of snags and logs.

Here we compare the beetle communities inhabiting snags and logs from base to crown to better understand the relative importance of these two habitats in southeastern forests. We predicted that overall species richness would be higher in logs than in snags based on previous research and on the idea that the upper bole sections and crowns of snags would be less accessible and therefore less readily colonized than those of logs.

Methods

Study site

This research took place on the 80,267-ha Savannah River Site (SRS) located in the upper Coastal Plain Physiographic Province of South Carolina. The SRS, a facility owned and operated by the United States Department of Energy, was established in 1951, and was designated a National Environmental Research Park in 1972 (Kilgo and Blake, 2005). Most of the land now owned by the Savannah River site was formerly used for agricultural purposes and most forests currently standing were planted or regenerated in the early 1950’s (Kilgo and Blake,

2005).

The SRS is somewhat typical of the southeastern coastal plain in that it is dominated

(68%) by pine forests growing on relatively dry upland sites and by mixed hardwoods (22%) occupying swamps and riparian bottomlands (Kilgo and Blake, 2005). However, the upland and bottomland sites do not consist purely of pines and hardwoods, respectively. At least three tree species are relatively common in both forest types. Sweetgum (Liquidambar styraciflua L.) and

7 water oak (Quercus nigra L.) grow most commonly on mesic sites dominated by mixed hardwoods but also appear sporadically among pines on dry upland sites. Similarly, loblolly pine (Pinus taeda L.) is currently the dominate pine species growing in upland pine forests but was historically restricted to moist bottomland sites (Schultz, 1997) and continues to grow there at low densities. Kilgo and Blake (2005) provide percent basal areas for tree species in different forest types on the Savannah River Site. For a shortleaf-loblolly pine slope, comparable to the upland forest used in this study, Pinus (taeda and echinata), L. styraciflua and Q. nigra made up

80%, 2% and 1% of the total basal area, respectively. In contrast, the average percent basal areas in bottomland forests bordering rivers and large streams for P. taeda, L. styraciflua and Q. nigra were 2.2%, 10.6% and 3.5%, respectively (Kilgo and Blake, 2005).

The upland and bottomland forests used in this study were approximately 25 km apart.

One Hobo Data Logger was placed in each forest type for approximately one year (2006-07) to record temperature and humidity. On average, the upland forest was warmer than the bottomland forest (18.8 and 17.8 ºC, respectively) whereas relative humidity was on average lower there than in the bottomland forest (72.2 and 76.6%, respectively). These differences were most pronounced during the growing season (Fig. 2.1).

Experimental design

Our sampling followed a 2 x 3 x 2 factorial design with the respective factors being forest type (upland pine forest vs. bottomland hardwood forest), tree species (L. styraciflua vs. P. taeda vs. Q. nigra), and posture (log vs. snag). There were three replicates.

On June 5-6 2006 we created 9 snags and 9 logs in the upland sites and the same number in the bottomland sites, equally divided among L. styraciflua, P. taeda, and Q. nigra (i.e., three

8 snags and logs of each species at each site). Snags were created by girdling the trees to a depth of 3 cm or more using a chainsaw and spraying full strength (53.8%) glyphosate (Foresters’®,

Riverdale Chemical Company, Burr Ridge, IL) into the wounds. To prevent the herbicide from traveling up the tree and possibly affecting insect colonization, a second girdle was created about

15 cm above the first before herbicide was applied. Only the lower girdle was treated. All girdled trees examined two weeks after treatment were dead.

Approximately 11 months later, in May 2007, we returned to collect sections from the three logs and snags of each species at each site. After felling the snags with chainsaws, we removed 0.5 m sections from the lower bole, middle bole, and upper bole of each snag and log.

The position of each section was measured from the tree base (Table 2.2). We also collected three 0.5 m crown sections taken from major limbs or sometimes the upper-most portion of the main bole. The tops of all but one of the sweetgum snags had broken, so those crown sections had been in contact with the ground for an unknown length of time. The upper bole sections from these trees were taken directly below the point of breakage. All the other snags were intact.

All bole and crown sections cut on a given day (May 3 and 8 for upland and bottomland forests, respectively) were labeled and transported to Athens, Georgia.

We recorded the diameter (measured at the center) and bark coverage (visual estimation) of each bole and crown section (Table 2.2) in the laboratory. We used these data to calculate the total surface area (not including ends) and bark surface area (product of surface area and visual estimate of bark coverage) sampled from each snag and log.

9 Insect rearing

Emerging beetles were collected in the laboratory using rearing bags. Rearing bags have been shown to be one of the most efficient methods for collecting saproxylic beetles from dead wood (Jonsell and Hansson, 2007). Bole (108) and crown sections (36) (i.e., the three branch sections from each tree were tied together) were suspended from wooden beams with synthetic rope and enclosed within large (170 l) extra-strength black plastic trash bags. In one bottom corner of each bag we attached a clear plastic collecting jar containing propylene glycol. To prevent mold problems, we continuously ventilated the bags using an electric blower (HADP9-1

Cast Aluminum Pressure Blower, Americraft Manufacturing Co., Cincinnati, Ohio). Air from the blower flowed through a plastic PVC pipe (~ 10 cm in diameter) that ran the length of the rearing facility near the ceiling. Each side of the pipe had rows of holes into which were inserted sections of clear vinyl tubing (0.95 cm OD, 0.64 cm ID). Each section of tubing led from the pipe to one of the rearing bags. The bags became inflated with air, thus forming effective funnels. Excess air escaped through a single small hole (~2 mm) drilled near the top of each collecting jar. Overhead florescent lights were left on at all times. We did not attempt to control temperature or humidity in the rearing facility, but all samples experienced the same conditions.

Screened windows were opened along both sides of the facility to allow for air movement and to match ambient conditions as closely as possible. However, it was typically warmer inside the facility than outside. Samples were collected about once a month for 20 weeks (4 May-21 Sept and 9 May-26 Sept for upland and bottomland samples, respectively) and transferred to 70% ethanol. Beetles were identified using the classification system of Arnett and Thomas

(2001,2002). Voucher specimens have been deposited in the Georgia Museum of Natural

History, Athens, Georgia.

10 Data analysis

To test whether there were any differences in the amount of surface area sampled, we conducted a three-way analysis of variance with total surface area sampled (summed for each log or snag) as the response variable. The analysis was repeated for total bark surface area sampled.

Bole and crown samples from each snag or log were combined before conducting an analysis of covariance on a three-way factorial design (SAS Institute, 1990). Surface area and bark surface area were the covariates and the main effects were forest type, tree species and wood posture. All effects were fixed and there were no missing or incomplete samples.

Species richness estimates, based on the Chao1 estimator, were calculated using

EstimateS (Colwell, 2006). The Chao1 estimator is calculated as follows: Chao1=Sobs + (a²/2b) where Sobs is the observed species richness, a is the number of singletons and b is the number of doubletons (Colwell and Coddington, 1994). This is an appropriate estimator for this study given that Chao 1 is thought to perform well on large datasets with large numbers of rare species

(Colwell and Coddington, 1994, and references therein). Species richness estimates are useful because, by factoring in species rarity, they give an indication of how thoroughly an assemblage of species has been sampled. Because it is possible for observed richness trends to differ from estimated richness trends, it is important to examine both.

Indicator species analysis (Dufrêne and Legendre, 1997) was performed four times using

PC-ORD (McCune and Mefford, 2006) to determine which species were significantly associated with 1) upland or bottomland forests; 2) snags or logs; 3) oak, pine or sweetgum; 4) lower bole, middle bole, upper bole or crown. Indicator values ranging from 0 (no association) to 100

(perfect association) were tested for statistical significance using a Monte Carlo randomization with 2500 permutations (McCune and Grace, 2002).

11 Results and Discussion

Data set

Overall, 33,457 specimens from 51 families and 250 “species” emerged over the 20 wk sampling period (Table 2.3). An effort was made to identify all specimens to the lowest taxonomic units possible given available time and expertise. All specimens were identified to family, 79% were identified to genus and 59% were identified to species. Several species rich groups (e.g., Ciidae, Corylophidae and Ptiliidae) were not sorted below family level and were treated as single taxonomic units even though they likely consisted of multiple species. The estimates of species richness presented in this paper are therefore conservative.

Surface area and bark surface area

Surface area did not vary significantly for any of the factors (data not shown). However, bark surface area varied significantly with treespecies (F2,24=31.30, P <0.0001), being lower for

P. taeda than for Q. nigra and L. styraciflua. There was also a significant interaction between treespecies and posture (F2,24=9.6, P=0.0009) due to the fact that P. taeda snags had considerably less bark than P. taeda logs (0.55 ± 0.12 and 1.22 ± 0.14 m², respectively).

Species richness and habitat associations

Overall, species richness differed significantly between forest types and wood postures but not among tree species (Table 2.1). Because there were no significant interaction terms

(Table 2.1), the results for each factor are discussed individually below.

1. Forest type. In total, 189 and 175 beetle species were collected from the upland and bottomland forests, respectively. Mean species richness was significantly higher in the upland

12 forest than the bottomland forest (Fig. 2.2a). We attribute this to differences in light intensity and temperature between the two forest types. The upland pine-dominated forest was more open and sun-exposed than the bottomland hardwood forest and was consequently warmer and less humid (Fig. 2.1). A number of studies have shown that sun-exposure promotes saproxylic beetle diversity (Bouget and Duelli, 2004, and references therein). For example, most saproxylic beetle species in Sweden, including 59% of those red-listed, can tolerate and often prefer sun-exposed conditions (Jonsell et al., 1998; Lindhe et al., 2005).

The two forest types supported fairly distinct communities even though we sampled the same tree species in both. Indicator species analysis determined that 15 and 9 species were significantly associated with the upland and bottomland forests, respectively (Table 2.3).

Further research is needed to better understand how and why saproxylic beetle communities differ between forest types. Fire frequency differs considerably between upland and bottomland forests and may be particularly important in shaping saproxylic beetle communities in the southeastern United States. For example, the frequent fires characteristic of upland forests may favor many pyrophilic species as they do in other regions (Evans, 1966;

Moretti et al., 2004). Also, frequent fires may select for enhanced dispersal abilities. Beetles in upland fire-prone forests may need to flee fires and re-colonize burned areas regularly compared to those in bottomland forests. This question has important implications with respect to the dead wood connectivity required in different forest types (Grove, 2006).

2. Tree species. There were no significant differences in beetle richness among tree species (Table 2.1). The observation that considerably fewer species emerged from P. taeda than L. styraciflua (Fig. 2.2b) may be attributed in part to the fact that bark surface area, a covariate in our model, was significantly lower for P. taeda than for L. styraciflua. However,

13 because 152 species emerged from L. styraciflua compared to just 122 and 125 species from Q. nigra and P. taeda, respectively, L. styraciflua may be of particular importance to saproxylic beetles in the southeastern United States.

The interaction between tree species and forest type was not significant (Table 2.1) even though tree abundances differed considerably between upland and bottomland forests. We expected more species would emerge from Q. nigra and L. styraciflua in the bottomland than in the upland forest because those species are much more common in bottomland forests.

Similarly, we expected P. taeda to support more species rich assemblages in the upland forest where that species is more abundant. The observed trends were not consistent with these expectations (Fig. 2.3). For example, Q. nigra yielded, on average, about 10 more beetle species in the upland pine-dominated stand than in the bottomland hardwood forest (Fig. 2.3). The expected species richness trends also did not follow the anticipated pattern (Fig. 2.4).

Recent findings from Germany corroborate our results. Müller and Goßner (2007) sampled saproxylic beetles in the crowns of oaks in both beech-dominated and oak-dominated forests. They found no difference in the proportion of oak specialists between forest types.

Furthermore, there was only a weak relationship between the proportion of oak specialists captured and surrounding oak density.

Our results show relative tree abundance is not a good predictor of beetle species richness and the seemingly minor hardwood components on upland sites are of considerable importance to the saproxylic beetle community. This may be particularly true for the hardwood-dominated drainages frequently embedded within upland pine stands in the southeastern United States.

These may be areas of high saproxylic beetle diversity and may provide refuge for saproxylic

14 beetles during fires. They might also greatly enhance habitat connectivity for species associated with hardwoods.

3. Wood posture. In total, 194 and 171 species emerged from logs and snags, respectively. Mean species richness was significantly higher in logs than in snags (Fig. 2.2c).

Similarly, species richness estimates were consistently higher for logs regardless of tree species and forest type (Fig. 2.4). These differences are consistent with previous studies (Jonsell and

Weslien, 2003; Gibb et al., 2006; McGeoch et al., 2007; Hjältén et al., 2007; Franc, 2007) and probably widen with time, particularly as the snags become dry following bark loss (Boulanger and Sirois, 2007).

Although snags support fewer beetle species than logs, it is clear from our results that a number of species specifically require snags. Using indicator species analysis, we found 12 species were significantly associated with snags and 18 species were significantly associated with logs (Table 2.3). A number of the snag-associated species were primarily collected from the upper-most portions of snags. For example, Tenebroides semicylindricus (Trogossitidae) was found to be significantly associated with the crowns of snags (Table 2.3). Similarly, almost all specimens of Germarostes (Ceratocanthidae) were collected from mid-bole or higher, including five specimens from crown sections. We also found evidence of vertical stratification among cossonine weevil genera. While the most common genus, Cossonus, was concentrated near the ground and was not significantly associated with snags, two other genera, Rhyncolus and

Stenoscelis, were significant snag associates and were collected most commonly from the upper- most bole sections (Fig. 2.5).

Based on our results and those of previous studies, snags appear vital to maintaining a complete saproxylic beetle community. Although logs support more species rich beetle

15 assemblages and have their own specialist species, our data and others suggest snags are more important than logs for conservation purposes. First, research from Scandinavia suggests that most saproxylic beetle species can live within standing dead wood and that snags support more threatened species than logs (Jonsell et al., 1998; Franc, 2007, and references therein). Second, snags become logs as soon as they fall, usually within 5 yrs for pine in the southeastern US

(Moorman et al., 1999; Conner and Saenz, 2005), thereby providing habitats for both snag and log-associated beetles. Third, logging slash, if left on site, should provide adequate habitat for many species associated with logs. Finally, snags are also required by a wide variety of cavity- nesting birds and other vertebrates of conservation concern (Lohr et al., 2002).

Acknowledgements

We wish to thank Scott Horn and Mike Cody for assisting with field work, Bob Rabaglia and

Alexey Tishechkin for identifying scolytines and histerids, respectively, and Cecil Smith and

Mike Thomas for their assistance and hospitality during visits to the Georgia Museum of Natural

History and the Florida State Collection of Arthropods, respectively. Support was provided by the Department of Energy-Savannah River Operations Office through the U.S. Forest Service

Savannah River under Interagency Agreement DE-AI09-00SR22188.

Literature Cited

Arnett, R.H., Jr., Thomas, M.C. (Eds.), 2001. American Beetles: Archostemata, Myxophaga,

Adephaga, Polyphaga: Staphyliniformia. CRC Press, Boca Raton, Florida.

Arnett, R.H., Jr., Thomas, M.C. (Eds.), 2002. American Beetles: Polyphaga: Scarabaeiodea

through Curculionoidea. CRC Press, Boca Raton, Florida.

16 Bouget, C., Duelli, P., 2004. The effects of windthrow on forest insect communities: a literature

review. Biol. Cons. 118: 281-299.

Boulanger, Y., Sirois, L., 2007. Postfire succession of saproxylic arthropods, with emphasis on

Coleoptera, in the north boreal forest of Quebec. Env. Ent. 36: 128-141.

Colwell, R.K., 2006. EstimateS: Statistical estimation of species richness and shared species

from samples. Version 8. Persistent URL .

Colwell, R.K., Coddington, J.A., 1994. Estimating terrestrial biodiversity through extrapolation.

Phil. Trans. R. Soc. Lond. B 345: 101-118.

Conner, R.N., Saenz, D., 2005. The longevity of large pine snags in eastern Texas. Wildl. Soc.

Bull. 33: 700-705.

Dufrêne, M., Legendre, P., 1997. Species assemblages and indicator species: the need for a

flexible asymmetrical approach. Ecol. Monograph. 67: 345-366.

Evans, W.G., 1966. Perception of infrared radiation from forest fires by Melanophila acuminata

De Geer (Buprestidae, Coleoptera). Ecology 47: 1061-1065.

Fonte, S.J., Schowalter, T.D., 2004. Decomposition in forest canopies. In: Lowman, M.D.,

Rinker, H.B. (Eds.), Forest Canopies. Elsevier Academic Press, Burlington,

Massachusetts. pp. 413-422.

Franc, N., 2007. Standing or downed dead trees- does it matter for saproxylic beetles in

temperate oak-rich forest? Can. J. For. Res. 37: 2494-2507.

Gibb, H., Pettersson, R.B., Hjältén, J., Hilszczański, J., Ball, J.P., Johansson, T., Atlegrim, O.,

Danell, K., 2006. Conservation-oriented forestry and early successional saproxylic

beetles: Responses of functional groups to manipulated dead wood substrates. Biol. Cons.

129: 437-450.

17 Grove, S.J., 2006. A research agenda for insects and dead wood. In: Grove, S.J., Hanula, J.L.

(Eds.), Insect biodiversity and dead wood: proceedings of a symposium for the 22nd

International Congress of Entomology. USDA For. Serv., Gen. Tech. Rep. SRS-93, pp.

98-108.

Hedgren, P.O., Schroeder, L.M., 2004. Reproductive success of the spruce bark beetle Ips

typographus (L.) and occurrence of associated species: a comparison between standing

beetle-killed trees and cut trees. For. Ecol. Manage. 203: 241-250.

Hjältén, J, Johansson, T., Alinvi, O., Danell, K., Ball, J.P., Pettersson, R., Gibb, H., Hilszczański,

J., 2007. The importance of substrate type, shading and scorching for the attractiveness of

dead wood to saproxylic beetles. Basic Appl. Ecol. 8: 364-376.

Jomura, M., Kominami, Y., Dannoura, M., Kanazawa, Y., 2008. Spatial variation in

respiration from coarse woody debris in a temperate secondary broad-leaved forest in

Japan. For. Ecol. Manage. 255: 149-155.

Jonsell, M., Weslien, J., Ehnström, B., 1998. Substrate requirements of red-listed saproxylic

invertebrates in Sweden. Biodiversity Conserv. 7: 749-764.

Jonsell, M., Weslien, J., 2003. Felled or standing retained wood- it makes a difference for

saproxylic beetles. For. Ecol. Manage. 175: 425-435.

Jonsell, M., Hansson, J., 2007. Comparison of methods for sampling saproxylic beetles in fine

wood. Entomol. Fenn. 18: 232-241.

Kappes, H., Topp, W., 2004. Emergence of Coleoptera from deadwood in a managed

broadleaved forest in central Europe. Biodiversity Conserv. 13: 1905-1924.

Kilgo, J.C., Blake, J.I. (Eds.), 2005. Ecology and Management of a Forested Landscape:

Fifty years on the Savannah River Site. Island Press, Washington, DC.

18 Larkin, P.A., Elbourn, C.A., 1964. Some observations on the fauna of dead wood in live

oak trees. Oikos 15: 79-92.

Lindhe, A., Lindelöw, Å., Åsenblad, N., 2005. Saproxylic beetles in standing dead wood

density in relation to substrate sun-exposure and diameter. Biodiversity Conserv. 14:

3033-3053.

Lohr, S.M., Gauthreaux, S.A., Kilgo, J.C., 2002. Importance of coarse woody debris to avian

communities in Loblolly pine forests. Conserv. Biol. 16: 767-777.

McCune B., Grace, J.B., 2002. Analysis of Ecological Communities. MjM Software Design,

Gleneden Beach, Oregon.

McCune, B., Mefford, M.J., 2006. PC-ORD. Multivariate Analysis of Ecological Data.

Version 5. MJM Software, Gleneden Beach, Oregon.

McGeoch, M.A., Schroeder, M., Ekbom, B., Larsson, S., 2007. Saproxylic beetle diversity in

a managed boreal forest: importance of stand characters and forestry conservation

measures. Diversity Distrib. 13: 418-429.

Moorman, C.E., Russell, K.R., Sabin, G.R., Guynn, D.C., Jr., 1999. Snag dynamics and

cavity occurrence in the South Carolina piedmont. For. Ecol. Manage. 118: 37-48.

Moretti, M., Obrist, M.K., Duelli, P., 2004. Arthropod biodiversity after forest fires: Winners

and losers in the winter fire regime of the southern Alps. Ecography 27: 173-186.

Müller, J., Goßner, M., 2007. Single host trees in a closed forest canopy matrix: a highly

fragmented landscape? J. Appl. Entomol. 131: 613-620.

Nordén, B., Götmark, F., Tönnberg, M., Ryberg, M., 2004. Dead wood in semi-natural

temperate broadleaved woodland: contribution of coarse and fine dead wood, attached

dead wood and stumps. For. Ecol. Manage. 194: 235-248.

19 Økland, B., 2002. Canopy cover favours sporocarp-visiting beetles in spruce forest.

Norw. J. Entomol. 49: 29-39.

SAS Institute, 1990. SAS Guide for Personal Computers. SAS Institute, Cary, NC, USA.

Schultz, R.P., 1997. Loblolly pine: The ecology and culture of loblolly pine (Pinus taeda

L.). Agricultural Handbook 713. USDA Forest Service, Washington, DC.

Stein, B.A., Kutner, L.S., Adams, J.S. (Eds.), 2000. Precious Heritage: The Status of

Biodiversity in the United States. Oxford University Press.

Sverdrup-Thygeson, A., Ims, R.A., 2002. The effect of forest clearcutting in Norway on the

community of saproxylic beetles on aspen. Biol. Cons.106: 347-357.

Warriner, M.D., Nebeker, T.E., Tucker, S.A, Schiefer, T.L., 2004. Comparison of

saproxylic beetle (Coleoptera) assemblages in upland hardwood and bottomland

hardwood forests. In: Spetich, M.A. (Ed.), Upland oak ecology symposium: history,

current conditions, and sustainability. USDA For. Serv., Gen. Tech. Rep. SRS-73, pp.

150-153.

20 Table 2.1. Results from an analysis of covariance on the three-way factorial design.

Source df MS F P forest type 1 185.8 4.62 0.0 3 4 tree species 2 112.0 2.78 0.0 3 8 wood posture 1 262.5 6.52 0.0 3 2 forest type*tree species 2 68.8 1.71 0.2 1 0 forest type*wood posture 1 30.84 0.7 0.3 7 9 treespecies*wood posture 2 15.86 0.3 0.6 9 8 forest type*tree species*posture 2 0.10 0.00 1.0 0 surface area (covariate) 1 117.6 2.92 0.1 7 0 bark surface area (covariate) 1 0.45 0.01 0.9 2 Error 2 40.26 2 Total 3 5

21 Table 2.2. Data (mean ± SE, n=3) collected from 0.5 m wood samples taken from 11 month-old logs and snags of three tree species

(Quercus nigra L., Pinus taeda L. and Liquidambar styraciflua L.) in South Carolina, USA. The samples were taken at four positions from each log and snag: lower bole, middle bole, upper bole and crown. Data from the three crown sections were summed. Surface area calculations do not include the ends of the logs. Bark surface area equals the product of surface area and % bark coverage (a visual estimate).

Distance from tree base (m) Diameter (m) Surface area (m²) Bark surface area (m²) Logs Snags Logs Snags Logs Snags Logs Snags Bottomland Q. n. lower 0.85 ± 0.23 0.68 ± 0.04 0.36 ± 0.01 0.35 ± 0.02 0.56 ± 0.02 0.55 ± 0.03 0.56 ± 0.02 0.55 ± 0.03 Q. n. middle 7.83 ± 0.14 8.42 ± 0.16 0.26 ± 0.01 0.27 ± 0.01 0.42 ± 0.02 0.42 ± 0.02 0.42 ± 0.02 0.42 ± 0.02 Q. n. upper 14.78 ± 0.34 15.33 ± 0.50 0.20 ± 0.00 0.20 ± 0.01 0.31 ± 0.01 0.31 ± 0.01 0.31 ± 0.01 0.29 ± 0.03 Q. n. crown . . 0.25 ± 0.02 0.24 ± 0.02 0.39 ± 0.03 0.38 ± 0.03 0.39 ± 0.03 0.38 ± 0.03 P.t. lower 0.52 ± 0.08 0.75 ± 0.14 0.36 ± 0.01 0.37 ± 0.04 0.57 ± 0.01 0.58 ± 0.06 0.54 ± 0.04 0 P.t. middle 9.97 ± 0.80 9.69 ± 0.84 0.28 ± 0.01 0.31 ± 0.04 0.43 ± 0.02 0.49 ± 0.06 0.42 ± 0.02 0.28 ± 0.17 P.t. upper 19.40 ± 0.95 18.11 ± 1.24 0.19 ± 0.01 0.22 ± 0.03 0.29 ± 0.01 0.35 ± 0.04 0.19 ± 0.09 0 P.t. crown . . 0.23 ± 0.01 0.21 ± 0.01 0.36 ± 0.01 0.33 ± 0.01 0.21 ± 0.04 0.10 ± 0.02 L.s. lower 0.99 ± 0.24 0.97 ± 0.04 0.32 ± 0.04 0.40 ± 0.02 0.50 ± 0.07 0.64 ± 0.03 0.50 ± 0.07 0.64 ± 0.03 L.s. middle 7.32 ± 0.12 9.11 ± 0.86 0.24 ± 0.02 0.29 ± 0.03 0.38 ± 0.04 0.46 ± 0.04 0.37 ± 0.04 0.40 ± 0.03 L.s. upper 14.17 ± 1.35 17.60 ± 1.83 0.19 ± 0.04 0.19 ± 0.01 0.30 ± 0.06 0.30 ± 0.01 0.29 ± 0.08 0.30 ± 0.01 L.s. crown . . 0.25 ± 0.03 0.23 ± 0.02 0.40 ± 0.04 0.36 ± 0.03 0.40 ± 0.04 0.32 ± 0.05 Upland Q. n. lower 0.66 ± 0.09 0.81 ± 0.14 0.33 ± 0.02 0.39 ± 0.04 0.51 ± 0.03 0.62 ± 0.07 0.38 ± 0.09 0.62 ± 0.07 Q. n. middle 4.84 ± 0.72 5.01 ± 0.15 0.26 ± 0.03 0.27 ± 0.02 0.41 ± 0.05 0.42 ± 0.03 0.37 ± 0.05 0.42 ± 0.03 Q. n. upper 8.78 ± 0.42 8.45 ± 0.38 0.19 ± 0.03 0.21 ± 0.02 0.30 ± 0.04 0.33 ± 0.04 0.24 ± 0.07 0.33 ± 0.04 Q. n. crown . . 0.24 ± 0.03 0.23 ± 0.02 0.38 ± 0.04 0.37 ± 0.04 0.36 ± 0.05 0.37 ± 0.04 P.t. lower 0.97 ± 0.38 0.73 ± 0.08 0.36 ± 0.01 0.35 ± 0.02 0.57 ± 0.02 0.55 ± 0.03 0.23 ± 0.12 0.38 ± 0.19 P.t. middle 8.27 ± 1.05 7.90 ± 0.32 0.31 ± 0.02 0.28 ± 0.01 0.49 ± 0.03 0.44 ± 0.02 0.25 ± 0.14 0.05 ± 0.05 P.t. upper 16.05 ± 1.78 16.00 ± 0.67 0.26 ± 0.02 0.23 ± 0.00 0.40 ± 0.03 0.35 ± 0.00 0.34 ± 0.04 0.14 ± 0.09 P.t. crown . . 0.20 ± 0.02 0.23 ± 0.01 0.31 ± 0.03 0.35 ± 0.02 0.28 ± 0.04 0.15 ± 0.05 L.s. lower 0.55 ± 0.03 0.83 ± 0.11 0.35 ± 0.01 0.31 ± 0.02 0.54 ± 0.02 0.48 ± 0.02 0.54 ± 0.02 0.48 ± 0.02 L.s. middle 5.24 ± 0.74 5.00 ± 0.28 0.26 ± 0.02 0.23 ± 0.01 0.40 ± 0.03 0.36 ± 0.02 0.40 ± 0.03 0.36 ± 0.02 L.s. upper 10.13 ± 1.43 8.88 ± 0.54 0.20 ± 0.02 0.18 ± 0.01 0.32 ± 0.03 0.28 ± 0.02 0.32 ± 0.03 0.28 ± 0.02 L.s. crown . . 0.22 ± 0.01 0.24 ± 0.01 0.35 ± 0.02 0.37 ± 0.02 0.35 ± 0.02 0.37 ± 0.02

22 Table 2.3. List of beetles collected from logs and snags of three tree species in two forest types at the Savannah River Site, South

Carolina, USA. Abundances are presented in terms of logs/snags. Associations are based on significant indicator values for 1) snags or logs; 2) upland or bottomland; 3) oak, pine, or sweetgum; and 4) bole position (lower, middle, upper or crown) with asterisks denoting significance: * P < 0.05, ** P < 0.01, *** P < 0.001.

Bottomland Upland Family/Species Association(s) Oak Pine Sweetgum Oak Pine Sweetgum Total (indicator value) Aderidae Cnopus impressus middle bole (17.8***) 0/0 5/3 0/0 0/0 3/0 1/0 12 (LeConte) pine (13.4**) Ganascus ptinoides 0/0 4/0 1/0 0/0 0/0 0/0 5 (Schwarz) Ganascus ventricosus 0/0 1/0 2/0 0/1 5/0 0/0 9 (LeConte) Anobiidae Lasioderma sp. 1/0 0/0 0/0 1/3 0/0 0/0 5 Petalium sp. bottomland (19.3**) 0/0 0/1 30/20 0/0 0/0 1/3 55 sweetgum (34.8***) Protheca sp. bottomland (9.4*) lower 0/0 0/0 15/45 0/0 0/0 2/0 62 bole (19.1***) sweetgum (16.7**) Tricorynus sp. middle bole (11*) 0/0 0/0 15/3 0/0 0/0 1/0 19 sweetgum (12.5**) Anthribidae Piesocorynus sp. oak (14.3**) log (9.5*) 7/0 0/0 0/1 38/0 0/0 0/0 46 Biphyllidae Diplocoelus rudis 0/0 0/0 0/0 1/0 2/0 3/0 6 (LeConte) Bostrichidae

23 Lichenophanes sp. 0/0 0/0 0/0 2/0 0/0 0/0 2 Bothrideridae Bothrideres geminatus snag (28.1**) 0/3 0/9 0/5 5/7 1/12 0/8 50 (Say) Prolyctus exaratus 0/2 0/0 0/0 0/0 0/0 0/0 2 (Melsheimer) Sosylus extensus Casey lower bole (18*) oak 73/84 0/0 0/11 60/91 0/0 7/11 337 (51.4***) Brentidae Arrenodes minutus 8/2 0/0 7/0 0/12 0/0 0/0 29 (Drury) Buprestidae Agrilus sp. oak (8.3*) 1/0 0/0 0/0 0/5 0/0 0/0 6 Buprestis lineata 0/0 0/2 0/0 0/0 0/0 0/0 2 Fabricius Chrysobothris femorata 2/0 0/0 0/0 1/0 0/0 0/1 4 Olivier Chrysobothris sexsignata 0/1 0/0 0/0 0/0 0/0 0/0 1 (Say) Carabidae Anillinus sp. 0/0 1/0 0/0 0/0 0/0 0/0 1 Coptodera aerata Dejean 0/1 0/0 0/0 0/0 0/0 0/0 1 Mioptachys flavicauda lower bole (45.3***) 8/0 39/23 36/94 10/0 70/5 49/6 340 (Say) log (29.4*) Perigona pallipennis 0/0 0/0 2/0 0/0 0/0 0/0 2 (LeConte) Phloeoxena signata 0/0 1/0 0/0 0/0 0/0 1/0 2 (Dejean) Polyderis laevis (Say) 0/0 0/0 0/0 0/0 1/0 0/0 1 Tachyta nana inornata upland (9.7*) log (7.9*) 0/0 0/0 0/0 3/0 12/1 2/0 18 (Say) Cerambycidae Acanthocinus nodosus lower bole (11.1*) pine 0/0 3/0 0/0 0/0 3/0 0/0 6 (Fabricius) (8.3*) Acanthocinus obsoletus pine (16.7**) 0/0 7/0 0/0 0/0 9/2 0/0 18 (Olivier)

24 Aegomorphus modestus sweetgum (8.3*) 0/0 0/0 0/0 0/0 0/0 10/0 10 (Gyllenhal) Aegomorphus 0/0 0/0 0/0 0/0 0/0 4/0 4 quadrigibbus (Say) Astylopsis sexguttata 0/0 1/0 0/0 0/0 0/0 0/0 1 (Say) Curius dentatus Newman 0/0 0/0 0/0 0/0 0/0 0/3 3 Elaphidion mucronatum 0/0 0/0 0/0 0/0 0/0 2/0 2 (Say) Leptostylus asperatus sweetgum (18.7***) 0/0 0/0 4/0 0/0 0/0 1/14 19 (Haldeman) Leptostylus planidorsus 0/0 0/0 0/0 0/0 0/0 1/0 1 (LeConte) Lepturges confluens sweetgum (8.3*) 0/0 0/0 0/2 0/0 0/0 0/2 4 (Haldeman) Liopinus alpha (Say) sweetgum (10.4*) 0/0 0/0 0/0 0/0 0/0 0/13 13 Monochamus carolinensis 0/0 2/0 0/0 0/0 0/1 0/0 3 (Olivier) Monochamus titillator middle bole (17.5*) 0/0 19/17 1/0 0/0 8/9 0/0 54 (Fabricius) pine (45***) Neoclytus scutellaris oak (18.7***) 2/2 0/0 0/0 2/9 0/0 0/0 15 (Olivier) Urographis fasciatus lower bole (26.6**) oak 74/113 0/0 36/31 20/61 0/0 42/45 422 (DeGeer) (43.7***) Xylotrechus colonus lower bole (30.1***) 7/12 1/0 4/7 3/78 0/0 13/11 136 (Fabricius) oak (23**) Xylotrechus sagittatus pine (61.9***) 0/0 20/41 1/0 0/0 14/32 0/0 108 (Germar) Ceratocanthidae Germarostes aphodioides oak (14.9**) snag 1/7 0/0 0/2 0/9 0/0 0/0 19 (Illiger) (11.8*) Germarostes globosus 0/4 0/0 0/0 0/0 0/0 0/0 4 (Say) Cerylonidae Cerylon unicolor (Ziegler) lower bole (23.7**) 1/6 1/5 19/37 0/0 5/8 12/34 128 sweetgum (18.3*)

25 Hypodacne punctata 0/0 0/0 1/0 0/0 0/0 0/0 1 LeConte Murmidius ovalis (Beck) upland (10.8*) oak 0/1 0/0 0/0 3/28 0/1 0/0 33 (16.2**) Mychocerinus depressus snag (8*) 0/22 0/0 0/0 0/0 0/2 1/1 26 (LeConte) Philothermus glabriculus log (12.5**) 0/0 0/0 4/0 1/0 11/0 6/0 22 LeConte Chelonariidae Chelonarium lecontei 0/0 0/0 0/4 0/0 0/0 0/0 4 Thomson Ciidae Ciidae spp. lower bole (46.5**) 2393/419 7/0 342/98 53/293 46/4 591/447 4693 Cleridae Ababa tantilla (LeConte) 1/0 0/0 0/0 0/0 0/0 0/0 1 Chariessa pilosa (Forster) 1/0 0/0 0/0 1/1 0/0 0/0 3 Cymatodera undulata 0/0 0/1 0/0 0/0 0/0 0/0 1 (Say) Neorthopleura thoracica oak (10.4*) crown 2/0 0/0 0/0 7/1 0/0 0/0 10 (Say) (10*) Priocera castanea 0/0 0/5 0/0 0/0 0/2 1/0 8 (Newman) Colydiidae Aulonium 0/0 0/0 0/1 0/5 0/0 0/1 7 parallelopipedum (Say) Bitoma carinata bottomland (32.1**) 124/74 11/4 25/100 15/28 8/0 3/32 424 (LeConte) lower bole (45***) oak (27.2**) Bitoma quadricollis oak (18.7***) log 9/0 3/0 0/0 14/3 0/0 0/0 29 (Horn) (12.5**) Bitoma quadriguttata upland (38.1***) oak 8/27 0/1 6/6 50/36 2/0 20/46 202 (Say) (32.4**) Colydium lineola Say lower bole (27.2*) oak 35/205 0/0 32/304 57/471 0/0 30/157 1291 (48.3***) snag (50.2**) Colydium nigripenne lower bole (19.3**) 0/0 5/47 0/0 0/0 55/22 1/0 130 LeConte pine (35.1***)

26 Endeitoma dentata (Horn) 0/0 0/0 0/0 0/0 1/0 0/0 1 Endeitoma granulata 0/0 0/1 0/0 0/0 13/0 0/0 14 (Say) Microsicus parvulus upland (9.7*) oak 0/0 0/0 0/0 2/9 0/0 0/0 11 (Guérin-Méneville) (14.6**) Namunaria guttulata 0/0 1/1 0/0 2/0 2/0 1/0 7 (LeConte) Nematidium filiforme oak (43.7***) snag 79/302 0/0 0/0 23/209 0/0 0/0 613 Leconte (17.4*) Synchita fuliginosa lower bole (40***) 26/56 1/0 12/68 59/37 0/1 183/15 458 Melsheimer sweetgum (32.9**) Corylophidae Corylophidae spp. oak (57.7***) log 460/2 3/2 72/4 980/69 6/3 36/1 1638 (48.8***) Cryptophagidae Atomaria sp. 0/0 1/0 0/1 0/0 0/0 0/0 2 Curculionidae Acalles minimus Blatchley 1/0 0/0 0/1 0/0 0/0 0/0 2 Caulophilus rufotestaceus 0/0 0/0 1/0 0/0 0/0 0/0 1 (Champion) Cossonus spp. pine (68.7***) 0/0 225/1255 0/0 0/1 630/1672 0/0 3783 Dryocoetes autographus 0/0 0/0 11/0 0/0 0/0 0/0 11 (Ratzeburg) Dryophthorus americanus 0/0 8/0 4/3 0/0 4/0 0/0 19 Germar Dryoxylon onoharaensum sweetgum (8.3*) 0/0 0/0 0/0 0/0 0/0 22/34 56 (Murayama) Euplatypus compositus upland (22.8*) middle 9/0 0/0 564/457 1588/169 0/0 381/437 3605 (Say) bole (23**) sweetgum (22.3*) Gnathotrichus 0/0 6/0 0/0 0/0 0/0 0/0 6 materiarius (Fitch) Himatium errans LeConte 0/0 0/1 0/0 0/0 1/2 0/0 4 Hypothenemus spp. upland (29.9**) lower 4/4 0/0 13/12 2/94 0/0 622/753 1504 bole (37.5***) sweetgum (38.8***)

27 Monarthrum mali (Fitch) 0/0 0/0 0/14 0/18 0/0 1/0 33 Myoplatypus flavicornis lower bole (13.9**) 0/0 167/1 0/0 0/0 65/0 0/0 233 (Fabricius) pine (10.4*) Oxoplatypus oak (53.6***) 761/966 0/0 38/1 171/1896 0/0 0/0 3833 quadridentatus Olivier Pityophthorus sp. 1 sweetgum (10.4*) 0/0 0/0 36/0 0/0 0/0 3/24 63 crown (10.9*) Pityophthorus sp. 2 0/0 0/4 0/0 0/0 0/1 0/0 5 Pityophthorus sp. 3 0/0 0/0 0/0 0/0 0/0 11/0 11 Pseudopentarthrum sp. 0/0 0/0 0/1 0/0 0/0 0/0 1 Pseudopityophthorus 0/0 0/0 0/0 1/5 0/0 0/0 6 pruinosus (Eichhoff) Rhyncolus sp. pine (29.2***) snag 0/0 7/55 0/0 0/0 1/355 0/0 418 (15**) Scolytus multistriatus 0/0 0/0 0/0 0/1 0/0 0/0 1 (Marsham) Stenoscelis andersoni bottomland (13.9**) 0/1 0/24 0/11 0/0 0/0 0/0 36 Buchanan snag (13.9**) Tomolips quercicola 0/2 0/2 0/3 0/3 0/0 0/0 10 (Boheman) Xyleborinus gracilis 0/0 0/0 0/0 0/25 0/0 0/0 25 (Eichhoff) Xyleborinus saxeseni sweetgum (22.8***) 0/0 0/0 441/36 48/1 0/0 25/15 566 (Ratzeburg) Xyleborus affinis Eichhoff lower bole (18.4*) 1/4 0/0 473/998 8/71 1/0 42/612 2210 sweetgum (36.1***) Xyleborus californicus 0/1 0/0 0/0 0/0 0/0 0/0 1 Wood Xyleborus ferrugineus lower bole (24.7**) 20/0 565/37 154/14 0/0 182/0 11/0 983 (Fabricius) pine (28.2***) log (26.3***) Xyleborus pubescens pine (18.7***) log 0/0 13/0 0/0 0/0 30/0 0/0 43 Zimmermann (12.5**) sp. 29 0/0 0/0 0/0 0/0 0/0 1/0 1 Dermestidae Trogoderma ornatum 0/0 0/1 0/0 0/0 0/0 0/0 1

28 (Say) Elateridae Ampedus luteolus (Say) 0/0 0/0 0/0 0/0 2/0 0/0 2 Dicrepidius ramicornis 0/0 1/0 0/0 0/0 1/0 0/2 4 (Palisot de Beauvois) Drapetes geminatus Say 0/0 0/0 0/0 0/0 0/0 1/0 1 Glyphonyx sp. 0/0 0/0 0/0 0/0 0/1 0/0 1 Endomychidae Clemmus minor (Crotch) 0/0 0/0 0/0 0/2 0/0 0/0 2 Micropsephodes sweetgum (10.4*) 0/0 0/0 1/19 0/0 0/0 0/0 20 lundgreni Leschen and Carlton Eucnemidae Dromaeolus sp. 0/0 1/0 0/0 0/0 0/0 0/0 1 Nematodes atropos Say 4/0 0/0 0/0 2/0 0/0 0/0 6 Histeridae Acritus exiguus (Erichson) sweetgum (21*) 3/7 0/0 1/100 26/2 0/0 16/15 170 Aeletes floridae (Marseul) 0/0 0/0 1/0 0/0 0/0 0/0 1 Aeletes politus (LeConte) 0/0 0/0 1/0 0/0 0/0 0/0 1 Aeletes simplex (LeConte) lower bole (15**) log 1/0 0/0 1/2 3/0 6/0 2/0 15 (13.2*) Bacanius punctiformis upland (22.4**) lower 2/0 1/2 20/7 15/0 32/21 17/47 164 (LeConte) bole (23.1**) Bacanius sp. 3 0/0 0/0 0/3 0/0 0/0 0/0 3 (undescribed) Bacanius tantillus 0/0 0/0 0/0 0/0 12/0 0/0 12 LeConte Baconia aeneomicans 0/0 0/0 0/0 0/0 1/0 1/0 2 (Horn) Eblisia carolina (Paykull) 0/0 0/0 1/0 0/0 1/0 0/0 2 Epierus regularis 0/0 0/0 0/4 0/1 0/0 0/1 6 (Palisot de Beauvois) Paromalus seminulum sweetgum (9.4*) 1/0 0/0 4/0 0/0 0/0 4/1 10 Erichson Platylomalus aequalis bottomland (8.3*) 1/0 0/0 4/3 0/0 0/0 0/0 8 (Say) sweetgum (9.1*)

29 Platysoma leconti sweetgum (18.3**) 2/0 0/1 4/9 4/0 0/0 1/5 26 Marseul Plegaderus transversus upland (11.9*) pine 0/0 0/5 0/0 1/0 7/22 0/0 35 (Say) (22.3***) Laemophloeidae Cryptolestes dybasi 0/0 0/0 0/0 5/1 0/0 0/0 6 Thomas Cryptolestes punctatus 0/0 0/0 0/0 3/0 0/1 5/0 9 (LeConte) Cryptolestes uncicornis oak (17.7**) 0/7 0/0 0/2 130/0 0/0 6/0 145 (Reitter) Laemophloeus biguttatus 0/0 0/0 0/0 4/1 0/0 1/0 6 (Say) Laemophloeus lower bole (13.1*) 2/6 0/0 34/0 1/0 0/0 65/0 108 megacephalus Grouvelle Lathropus vernalis upland (27.9**) oak 4/75 0/0 1/2 6/538 1/2 7/4 640 LeConte (42.6***) snag (32.3***) Leptophloeus angustulus 1/0 0/0 0/0 1/1 0/0 0/0 3 (LeConte) Narthecius grandiceps 0/0 0/0 0/0 0/1 0/0 0/0 1 LeConte Phloeolaemus 0/0 0/0 5/0 1/0 0/0 0/0 6 chamaeropis (Schwarz) Placonotus modestus 0/0 0/0 0/0 2/0 0/0 0/1 3 (Say) Placonotus zimmermanni oak (30.7***) log 13/0 0/0 0/0 39/1 0/0 1/0 54 (LeConte) (20.4***) Latridiidae Cartodere constricta 0/0 0/0 0/0 0/1 0/1 0/0 2 (Gyllenhal) Corticarina sp. 0/0 0/0 0/0 0/0 1/0 6/0 7 Enicmus sp. 0/0 0/0 0/0 0/3 0/0 0/0 3 Leiodidae Agathidium sp. 0/0 0/0 0/0 0/0 0/0 1/0 1 Lycidae

30 Plateros sp. 0/0 0/0 0/0 0/0 0/0 1/0 1 Melandryidae Phloeotrya sp. 1/0 0/0 0/0 0/1 0/0 1/0 3 Melyridae Attalus sp. 0/0 0/0 0/0 0/4 0/0 1/0 5 Micromalthidae Micromalthus debilis bottomland (8.3*) 1/1 0/0 2/2 0/0 0/0 0/0 6 LeConte Monotomidae Bactridium sp. upland (12.5**) oak 0/0 0/0 0/0 23/2 0/0 1/0 26 (16***) Monotoma sp. 0/0 0/0 0/0 0/1 0/0 0/0 1 Rhizophagus sp. 0/0 0/0 0/0 0/0 1/0 0/0 1 Mordellidae sp. 1 0/0 0/0 2/0 0/0 0/0 0/0 2 sp. 2 0/0 0/0 2/0 0/0 0/0 0/0 2 Mycetophagidae Litargus sexpunctatus upland (9.7*) oak 0/0 0/0 0/0 5/10 0/0 0/0 15 (Say) (14.6**) Litargus sp. 2 0/0 0/0 0/0 1/0 0/0 0/0 1 Mycetophagus pini 0/0 2/0 0/0 0/0 1/0 0/0 3 Ziegler Thrimolus minutus Casey 0/0 0/0 0/1 0/0 1/0 0/0 2 Nitidulidae Carpophilus sp. 1 0/0 0/0 0/0 0/0 1/0 0/0 1 Carpophilus sp. 2 1/1 0/0 0/0 0/2 0/0 0/0 4 Epuraea luteolus 0/0 0/0 0/0 1/1 0/0 1/0 3 (Erichson) Prometopia sexmaculata upland (21.2***) oak 4/0 0/0 2/0 32/8 4/1 7/2 60 (Say) (21.4***) log (17*) Passandridae Catogenus rufus 0/0 0/0 0/0 0/0 0/1 0/0 1 (Fabricius) Phalacridae sp. 1 0/43 0/0 0/0 0/0 0/0 0/0 43 sp. 2 0/0 0/0 0/0 0/0 0/0 1/0 1

31 Ptiliidae spp. pine (28.1***) 1/0 32/222 3/0 0/0 14/4 1/5 282 Pyrochroidae Dendroides canadensis 0/0 0/0 0/0 0/0 0/0 3/0 3 LeConte Rhysodidae Omoglymmius 0/0 0/0 1/0 0/0 0/0 0/0 1 americanus (Laporte) Scirtidae Cyphon sp. 0/0 0/0 1/1 0/0 0/0 0/0 2 Scraptiidae Canifa sp. 0/0 0/0 0/0 0/0 0/5 0/0 5 Scydmaenidae sp. 1 0/0 0/0 0/0 0/0 1/0 0/0 1 sp. 2 0/0 0/0 0/0 0/0 0/1 0/0 1 Silvanidae Ahasversus advena 3/1 2/0 0/2 1/15 0/1 0/1 26 (Waltl) Cathartosilvanus imbellis 0/0 1/0 0/5 14/3 0/0 5/0 28 (LeConte) Silvanus muticus Sharp upland (9.9*) 0/0 0/0 1/0 2/0 2/0 1/3 9 Silvanus planatus Germar 0/0 0/0 0/0 6/0 0/0 2/0 8 Sphindidae Sphindus sp. 0/0 1/0 0/0 0/0 1/0 4/0 6 Staphylinidae Anacyptus testaceus lower bole (9.4*) pine 0/0 1/1 0/0 0/0 36/0 0/0 38 (LeConte) (10.4**) Clavilispinus sp. lower bole (41.3***) 38/99 124/10 28/153 14/40 90/0 27/79 702 Hesperus sp. 0/0 0/0 0/2 0/0 0/0 0/0 2 Homaeotarsus sp. 0/0 0/0 1/0 0/0 0/0 0/0 1 Myrmecocephalus sp. lower bole (11.1*) 0/0 3/1 0/0 0/6 0/0 0/1 11 Myrmecosaurus 0/0 0/0 0/0 0/0 1/0 0/0 1 ferrugineus Bruch Scaphisoma sp. 0/0 0/0 0/0 0/0 0/0 2/0 2 Sunius sp. 0/0 0/0 0/0 0/0 0/0 1/0 1 Thoracophorus costalis log (35.8***) 5/0 4/0 72/0 28/0 34/1 83/10 237

32 (Erichson) Toxidium sp. 0/0 0/0 0/1 0/0 0/0 0/0 1 sp11 0/0 1/0 0/0 1/0 0/0 0/0 2 sp12 0/0 0/0 0/3 3/0 0/0 0/0 6 sp13 0/0 0/0 0/0 1/0 0/0 1/0 2 sp14 0/0 0/0 0/0 0/0 0/0 0/1 1 sp15 5/0 0/0 0/0 0/0 0/0 0/0 5 sp16 bottomland (8.3*) 0/1 0/3 7/3 0/0 0/0 0/0 14 sp17 0/0 0/2 0/0 0/0 0/0 0/0 2 sp18 0/0 0/1 0/0 0/0 0/0 0/0 1 sp19 0/0 0/2 0/0 0/0 0/0 0/0 2 sp20 0/0 1/6 0/0 0/0 0/0 0/0 7 sp21 0/0 0/1 0/0 0/0 0/0 0/0 1 sp22 0/0 0/1 0/0 0/0 0/0 0/0 1 sp23 0/0 1/0 0/0 0/0 0/0 0/0 1 sp24 pine (18.7***) log 0/0 3/2 0/0 0/0 25/0 0/0 30 (9.1*) sp25 0/0 0/1 0/0 0/0 0/0 0/0 1 sp26 0/0 0/1 0/0 0/0 0/0 0/0 1 sp27 pine (8.3*) 0/0 1/0 0/0 0/0 3/0 0/0 4 sp28 sweetgum (15.5*) 0/0 0/0 83/36 0/0 4/6 9/0 138 sp29 0/0 0/0 1/0 0/0 0/0 0/0 1 sp30 0/0 0/0 1/0 1/0 0/0 0/0 2 sp31 0/0 0/0 14/0 0/0 0/0 0/0 14 sp32 0/0 0/0 0/1 0/0 0/0 1/0 2 sp33 0/0 0/0 0/0 0/0 2/0 0/0 2 sp34 0/0 0/0 0/0 0/0 0/0 0/4 4 sp35 0/0 0/0 0/0 0/0 0/0 1/0 1 sp36 0/0 0/0 0/0 0/0 1/0 0/0 1 sp37 0/0 0/0 0/0 0/0 0/1 0/0 1 sp38 0/0 0/0 0/0 0/0 2/0 0/0 2 sp39 0/0 0/0 0/0 0/0 1/0 0/0 1 sp40 0/0 0/0 0/0 0/0 2/0 0/0 2 sp41 0/0 2/1 1/0 0/0 0/1 0/0 5 sp42 0/0 0/0 0/0 1/0 0/0 0/0 1 sp43 0/0 0/0 0/0 0/0 0/2 0/0 2

33 sp44 0/0 0/0 1/1 0/0 0/0 0/0 2 sp45 0/0 0/0 0/0 0/0 0/0 1/0 1 sp46 0/0 0/0 0/1 0/0 0/0 0/0 1 sp47 0/0 0/0 0/0 0/0 1/0 0/0 1 sp48 0/0 0/1 0/0 0/0 0/0 0/0 1 sp49 0/0 0/0 0/1 0/0 0/0 0/0 1 Synchroidae Synchroa punctata 1/0 0/0 0/0 0/0 0/0 0/0 1 Newman Tenebrionidae Adelina pallida (Say) 0/0 0/0 0/0 0/0 0/0 0/3 3 Alobates pennsylvanicus lower bole (9.9*) 0/0 0/0 3/1 0/1 0/1 0/1 7 (Degeer) Corticeus thoracicus lower bole (24.2**) 3/0 55/6 27/113 0/4 76/0 0/2 286 (Melsheimer) Gnathocerus maxillosus 0/0 0/0 0/0 0/1 0/0 0/0 1 (Fabricius) Hymenorus sp. 0/0 0/0 15/0 0/0 0/0 0/0 15 Liodema laeve lower bole (16.7***) 0/1 0/0 0/16 0/1 0/0 0/1 19 (Haldeman) snag (8.3*) Lobopoda erythrocnemis 0/0 0/0 0/0 0/0 1/0 0/0 1 Germar Platydema excavatum 0/0 0/0 0/0 1/0 0/0 0/0 1 (Say) Platydema flavipes pine (13.6*) log 0/0 13/0 2/0 0/0 14/0 2/2 33 (Fabricius) (14.4**) Platydema picilabrum 0/1 0/0 0/0 0/0 0/0 0/0 1 Melsheimer Platydema ruficorne lower bole (13.6*) pine 4/0 142/0 0/0 0/0 59/2 10/0 217 (Stürm) (11.7*) log (11*) Platydema subcostatum 1/0 0/0 0/0 0/0 0/0 1/0 2 Laporte and Brulle Poecilocrypticus 0/0 0/0 0/0 0/0 0/1 0/0 1 formicophilus Gebien Tetratomidae Eustrophus tomentosus 0/0 0/0 1/0 0/0 0/0 0/0 1

34 Say Throscidae sp. 0/0 1/0 0/0 0/0 0/0 0/0 1 Trogossitidae Airora cylindrica pine (19.9***) snag 0/0 1/13 0/1 0/2 1/11 0/1 30 (Serville) (16.9**) Corticotomus cylindricus 0/0 0/0 0/0 0/0 0/1 0/0 1 (LeConte) Lycoptis americana 0/0 0/0 0/0 0/0 0/1 0/0 1 (Motschulsky) Temnoscheila virescens sweetgum (41.2***) 5/3 0/1 16/19 2/8 1/2 8/10 75 (Fabricius) Tenebroides bimaculatus 0/4 0/0 0/0 0/0 0/0 0/0 4 (Melsheimer) Tenebroides collaris 0/0 1/1 0/0 0/0 0/2 0/0 4 (Sturm) Tenebroides corticalis lower bole (18.6*) log 5/2 18/0 7/3 10/0 2/0 9/10 66 (Melsheimer) (25.8**) Tenebroides laticollis oak (16.3**) 1/1 0/0 0/0 1/38 0/0 1/0 42 (Horn) Tenebroides marginatus upland (9*) lower bole 0/0 0/1 0/0 6/0 2/5 0/0 14 (Palisot de Beauvois) (10.9*) pine (8.3*) Tenebroides nanus bottomland (12.4*) oak 0/30 0/0 0/3 0/0 0/0 0/4 37 (Melsheimer) (11.8*) snag (16.7**) Tenebroides snag (14.1**) crown 0/2 0/4 0/0 1/2 0/0 0/4 13 semicylindricus (Horn) (12*) Thymalus marginicollis 0/0 0/0 0/0 0/0 0/0 1/0 1 Chevrolat Zopheridae Hyporhagus punctulatus 0/0 0/0 0/0 1/0 0/0 3/0 4 Thomson Pycnomerus haematodes upland (13.2**) pine 0/0 0/4 0/0 0/0 11/64 1/0 80 (Fabricius) (20.6***) Pycnomerus reflexus bottomland (19.4***) 28/0 5/0 22/3 0/0 0/0 0/0 58 (Say) lower bole (12.3*) log (14.5**)

35 Pycnomerus sulcicollis lower bole (20.3**) 2/0 6/0 3/0 3/0 33/0 1/0 48 LeConte pine (15.2*) log (22.2***) Total no. of individuals 4262/2608 1552/1837 2799/2918 3633/445 1683/2297 2470/2946 3345 2 7 Number of species, total 58/45, 76 56/51, 83 72/66, 100 72/65, 97 71/46, 92 85/53, 105 250

36 Fig. 2.1. Temperature and relative humidity over time for a bottomland hardwood forest and an upland pine-dominated stand in South Carolina, USA.

37 RH (%) Temp (C) 10 15 20 25 30 35 55 60 65 70 75 80 85 0 5

Nov 22

Dec 13 Upland Bottomland Jan 3 Jan 24 Feb 14 Mar 7 Date (2006-07) Date Mar 28 Apr 18

38 May 9 May 30 Jun 20 Jul 11 Aug 1 Aug 22 Sep 12 Oct 3 Oct 24 Fig. 2.2. Mean (± SE) number of saproxylic beetle species from two forest types (A), three tree species (B) and two wood postures (C) in South Carolina, USA. The P-values are based on an analysis of covariance for the three-way factorial design (Table 1).

39 No. of Species 25 30 35 40 45 A) P =0.04 Forest type Forest

Bottomland

Upland (n=18) B) P =0.08 Tree species Tree

Q. nigra 40

P. taeda

L. styraciflua (n=12) C) P =0.02 Wood posture posture Wood

Log

Snag (n=18) Fig. 2.3. Mean (± SE) number of saproxylic beetle species collected from wood in upland and bottomland forests for each of three tree species in South Carolina, USA.

41 50

Bottomland Upland 45

40

35

30

Mean Number of Species (n=6) 25

Q. nigra P. taeda L. styraciflua Tree Species

42 Fig. 2.4. Lower dots indicate observed total numbers of beetle species collected from ~11 month- old logs and snags of three tree species (Q. nigra, P. taeda and L. styraciflua) in South Carolina,

USA. Above these are the mean (n=3) Chao1 species richness estimates with 95% confidence limits. Sampling took place in both a mixed bottomland hardwood forest (left) and an upland pine-dominated stand (right).

43 350 Bottomland Upland 300

250

200

150

100

Numberof BeetleSpecies 50

0

Q.n. log P. t. log L. s. log P. t. log L. s. log P. t. snag L. s. snagQ. n. log Q. n. snag Q. n. snag P. t. snag L. s. snag

44 Fig. 2.5. Vertical distribution patterns of the three most common Cossoninae (Curculionidae) genera collected from pine snags in South Carolina, USA.

45 500 Cossonus 400 300 200 100 0 Rhyncolus 60

40

20

0 Stenoscelis 5 4

Mean No. of Individuals (n=6) 3 2 1 0 LOWER MIDDLE UPPER (0.7 ± 0.1) (8.8 ± 0.5) (17.1 ± 0.7) Bole Position (mean height (m), n=6)

46 CHAPTER 3

RESPONSES OF ARTHROPODS TO LARGE-SCALE MANIPULATIONS OF DEAD

WOOD IN LOBLOLLY PINE STANDS OF THE SOUTHEASTERN UNITED STATES¹

¹Ulyshen, M.D. and J.L. Hanula. Submitted to Environmental Entomology,

12/30/2008.

47 Abstract

Large-scale experimental manipulations of dead wood are needed to better understand its importance to animal communities in managed forests. In this experiment, we compared the abundance, species richness and diversity of arthropods in plots in which either 1) all coarse woody debris was removed, 2) a large number of logs were added, 3) a large number of snags were added, or 4) no coarse woody debris was added or removed. The target taxa were ground- dwelling arthropods, sampled by pitfall traps and saproxylic beetles (i.e, dependent on dead wood), sampled by flight intercept traps and emergence traps. There were no differences in total ground-dwelling arthropod abundance, richness or diversity among treatments. Only the results for ground beetles (Carabidae), which were more species rich and diverse in log input plots, supported our prediction that ground-dwelling arthropods would benefit from additions of dead wood. There were also no differences in saproxylic beetle abundance, richness or diversity among treatments. The results from this study are encouraging in that arthropods appear less sensitive than expected to manipulations of dead wood in managed pine forests of the southeastern United States. Because log input plots and removal plots supported similarly diverse communities of arthropods, we cannot recommend inputting large amounts of dead wood for conservation purposes, given the expense of such measures. However, the persistence of saproxylic beetles requires that an adequate amount of dead wood is available in the landscape and we recommend that dead wood be retained whenever possible in managed pine forests.

Keywords: coarse woody debris, biodiversity, epigaeic arthropods, forest management, conservation, colonization, dispersal

48 Introduction

Since the first forests appeared on Earth over 355 million years ago (Scheckler, 2001), animal communities have been developing and diversifying in the presence of dead wood. The extent to which species have come to rely, either directly or indirectly, on this important resource ranges from no association to complete dependence (i.e., “saproxylic”). Many species appear to fall somewhere between these two extremes, benefiting in some way from the presence of dead wood but not requiring it. Previous studies have shown that 1) 20-25% of all forest-dwelling species depend on dead wood, 2) managed forests have less dead wood than unmanaged forests and 3) many saproxylic species have become threatened in intensively managed landscapes (e.g.,

Siitonen, 2001; Grove, 2002; Langor et al., 2008). Large-scale experiments are needed to determine how forest communities respond to changing amounts of dead wood in managed forests (Davies et al., 2008). Such information is critical if we hope to satisfy both timber demands and conservation goals in the long term. Here we present results from a study examining the responses of arthropods to large-scale manipulations of dead wood in loblolly pine (Pinus taeda L.) forests of the southeastern United States.

This research is part of a multidisciplinary effort to study the responses of animals to the addition and removal of coarse woody debris in loblolly pine forests on the Savannah River Site,

South Carolina. In this experiment, we compared the abundance, species richness and diversity of arthropods in plots in which either 1) all coarse woody debris was removed, 2) a large number of logs were added, 3) a large number of snags were added, or 4) no coarse woody debris was added or removed (i.e., for reference). The target taxa were 1) ground-dwelling arthropods, sampled by pitfall traps and 2) saproxylic beetles, sampled by flight intercept traps and emergence traps.

49 A large number of studies have shown that many ground-dwelling arthropod taxa respond positively to dead wood at small-scales (e.g., Buddle, 2001; Varady-Szabo and Buddle, 2006;

Nittérus and Gunnarsson, 2006; Ulyshen and Hanula, 2009a, and references therein). However, the extent to which manipulations of dead wood over large areas affect the abundance and diversity of these taxa remains largely unknown. There have been too few large-scale experiments to adequately address this question, although several studies suggest that beetles

(especially ground beetles) (Cárcamo and Parkinson, 1999; 2003; Gunnarsson et al., 2004; Latty et al., 2006; Hanula et al., 2006) and spiders (Hanula et al., 2006) respond positively to dead wood at large scales. One of these studies took place in the reference and removal plots used in this project, soon after the plots were established in 1996. Hanula et al. (2006) sampled ground dwelling arthropods in those plots for five years (1997-2001) and found no differences in overall abundance or morphospecies richness. However, overall arthropod diversity and evenness were significantly (α=0.1) lower in removal plots than in reference plots and several families differed in abundance between the two treatments. These differences were observed only in the first two full years of sampling (i.e., 1998 and 1999), however. In this study, we sampled ground- dwelling arthropods using pitfall traps for four more years in the same reference and removal plots as well as the log input and snag input plots established in 2001. We predicted that arthropods overall and many individual orders and families would become more abundant, rich and diverse in response to the addition of dead wood.

Beetles are the most conspicuous and diverse inhabitants of dead and dying wood. An estimated 20-25% of all beetle species are thought to be saproxylic (Elton, 1966; Grove, 2002) and many of those (e.g., about 40% in Sweden) have become threatened in intensively managed landscapes (Jonsell et al., 2004, and references therein). Most evidence of imperilment comes

50 from the boreal forests of Scandinavia whereas the status of saproxylic beetles in other regions, including the southeastern United States, remains largely unknown. Saproxylic beetles were sampled in this study using flight intercept traps and emergence traps to determine how they differed in abundance, richness and diversity among treatments. We predicted that saproxylic beetles would become more abundant, rich and diverse in response to the addition of dead wood.

Materials and Methods

Study Site

This research took place on the 80,267-ha Savannah River Site (SRS) located on the upper Coastal Plain Physiographic Province of South Carolina. The SRS, a facility owned and operated by the United States Department of Energy, was established in 1951, and was designated an Environmental Research Park in 1972 (Kilgo and Blake, 2005). Innumerable studies have since been conducted to better understand the environmental impacts of human activities on forest ecosystems. Most of the land now owned by the Savannah River site was formerly used for agricultural purposes and most forests currently standing, including those used in this study, were planted in the early 1950’s (Kilgo and Blake, 2005).

Stand characteristics and site preparation

The loblolly pine (Pinus taeda L.) stands used in this study were planted between 1950 and 1953. Loblolly pine dominates much of the SRS, and constitutes one of the most economically important forest types in the southeastern United States (Schultz, 1997). Water oak (Quercus nigra L.), sweetgum (Liquidambar styraciflua L.) and several less common tree species were also present in low numbers. The stands were all thinned between 1991 and 1996

51 to achieve a standing basal area of 13.8-20.8 m²/ha (McCay et al., 2002). The understory varied somewhat but was generally dominated by wax myrtle (Myrica cerifera L.), blackberry (Rubus spp.), kudzu (Pueraria montana (Lour.) Merr.), Lespedeza bicolor Turcz., and Japanese honeysuckle (Lonicera japonica Thunb.). To help control for differences in plant cover, all stands were treated with herbicide in 1996.

Fire history differed somewhat among the plots. Although most plots had been burned between 1990 and 1994, others had not been burned since 1983, or as early as 1972. To help control for differences in fire history, prescribed fires were administered in all plots between

February 2000 and March 2001.

Experimental design

In this randomized complete block design, four blocks were selected and divided into four square 9.3 ha plots (Figure 3.1). Each plot consisted of a 6-ha core surrounded by a 3.3-ha buffer area (McCay et al., 2002). Each of the four plots within each stand was randomly assigned to one of the following treatments (Figure 3.1), as outlined by Moseley et al. (2008).

1) Removal: Removal of all dead woody material, including snags, ≥ 10 cm in diameter and

≥ 60 cm in length. This began in 1996 and was repeated yearly for the duration of the

study. All removed material was dumped in designated piles outside the plots.

2) Log input: 5-fold increase in log volume over average background levels. Logs were

added in 2001 by felling trees within the plots.

3) Snag input: 12-fold increase in snag basal area over the average snag basal area on the

plots before treatment. Snags were created by girdling and herbicide treatment in 2001.

52 4) Reference: aside from being thinned between 1991 and 1996 (see above), the reference

stands were not manipulated and were comparable to the forest matrix surrounding the

plots.

Pitfall trapping

Three rows of five pitfall traps were arranged in a grid-like pattern in the center (6-ha core area) of each plot for a total of 15 traps. The traps were spaced 50 m apart and were therefore independently placed with respect to dead wood and other features of the forest floor.

Each trap consisted of a 480 ml plastic cup buried to ground level and was positioned at the intersection of four 0.5 meter-long drift fences. A small funnel (8.4 cm diameter) inserted into the cup directed captured arthropods into a smaller 120 ml specimen cup below, which contained a 1% formaldehyde solution for specimen preservation. The traps ran for a week, and we sampled every other month for four years beginning in March, 2002, and ending in December,

2005. Samples from the fifteen traps in each plot were combined in the field and stored in 70% ethanol. Four laboratory technicians separated arthropods from the samples and sorted them by size and shape. Because samples from different sampling periods were processed by different people, we cannot meaningfully compare sampling periods and do not include time in our model

(i.e., we combined all sampling periods prior to analysis). Samples were then further sorted to morphospecies (MDU) using an established reference collection. Mites, Collembola and other micro-arthropod taxa were not counted and holometabolous insect larvae were excluded from the dataset. Analyses of variance were performed using SAS to determine whether there were any differences in abundance, species richness or Shannon’s diversity among treatments.

53 Flight intercept trapping

A pair of flight intercept traps (Ulyshen and Hanula, 2007) were placed at the center of each reference, removal, and log input plot. The traps were placed 10 m apart, and were suspended about 0.5 m above the ground from metal poles. Propylene glycol was used as a preservative and samples were collected every two weeks. The traps ran continuously for six weeks in 2007 (April 26-June 7). Specimens were sorted by species or morphospecies and those that are known to live in loblolly pine (i.e., based on emergence data collected in this and related studies) were categorized as “saproxylic”. Data from the two traps in each plot and the different sampling periods were combined before analysis. Analyses of variance were performed using

SAS to determine whether there were any differences in abundance, species richness or

Shannon’s diversity among treatments.

Emergence trapping

On May 4-5 2006 we cut 180 0.5 m sections (mean diameter 29.8 ± 0.3 cm, range 23.6-

37.3 cm) of loblolly pine from nine freshly felled trees. The lower 2 meters were discarded to minimize differences in bark thickness. Five logs were placed in a circular arrangement (Figure

3.2) at the edge, center, and half-way between the edge and the center of log removal, log input and reference plots (Figure 3.1). Care was taken to ensure that logs placed on the edges of plots were not adjacent to other treatments (Figure 3.1). The five logs in each group were randomly assigned a number (1-5). Two months after cutting (5 July) we returned to collect log number 1 from each location. Log numbers 2, 3 and 4 were collected after about 6 (12 November), 10 (17

March), and 22 months (7 March), respectively. The fifth log from each group was not collected. Logs were loaded onto trucks and transported to an emergence facility in Athens,

54 Georgia. Although the logs were handled as carefully as possible, some bark loss or loosening occurred. Loose bark was reaffixed to logs with string or wire, when possible. Beetles emerging from each set of logs were collected for 6 months using rearing bags (Ulyshen and Hanula,

2009b) and later identified by MDU.

Sampling followed a split-split plot design consisting of three treatments (i.e., whole plots: log removal, log input and reference), three locations (i.e., split plots: edge, center, and half-way between edge and center), and four dates (i.e., split-split plots: 2, 6, 10 and 22 months).

Analyses of variance were performed using SAS with species richness as the response variable.

The same analysis was performed on abundance data for the 25 most numerous (i.e., >100 individuals) taxa. There were no missing or incomplete samples.

Results

Pitfall trapping

Overall, 210 656 specimens representing 1206 morphospecies were collected by pitfall traps

(Table 3.1). There were no significant differences in abundance, species richness or Shannon’s diversity among treatments for arthropods overall and only five significant differences for the individual taxa examined (Table 3.2). The beetle family Carabidae was significantly more species rich and diverse in log input plots than in the other treatments (Table 3.2, Table 3.6).

Flight intercept trapping

The flight intercept traps captured 11 600 beetle specimens representing ≥ 240 species overall. Of these, 1769 (15%) specimens and 60 (25%) species are known to be saproxylic on

55 loblolly pine (Table 3.7). There were no differences in species richness or diversity among treatments for beetles overall or for saproxylic species alone (Table 3.3).

Emergence trapping

A total of 16 347 beetle specimens representing ≥ 155 species emerged from the 144 logs sampled in this study (Table 3.8). Species richness did not differ among treatments or among locations and there were no significant interactions (Table 3.4). However, species richness did differ significantly among sampling dates, increasing with log age (mean ± SE richness (n=36):

12.3 ± 0.6, 11.6 ± 0.5, 15.7 ± 0.9, 18.1 ± 1.1 for 2, 6, 10 and 22 month-old logs, respectively).

Similarly, none of the most common taxa differed in abundance among treatments or locations

(Table 3.5) but most differed in abundance among sampling dates. There were significant interaction terms for some species, but no more than expected by chance.

Discussion

1. Ground-dwelling arthropods. There were no differences in total arthropod abundance, richness or diversity among treatments. For the 73 comparisons made on individual taxa, only five revealed significant differences, about the number expected by chance at the α=0.05 level of significance. Only the results for ground beetles (Carabidae), which were more species rich and diverse in log input plots, support our prediction that arthropods would become more abundant, species rich or diverse in response to additions of dead wood. These results are consistent with previous studies that have shown positive associations between ground beetles and coarse woody debris (Cárcamo and Parkinson, 1999; Pearce et al., 2003; Latty et al., 2006; Nittérus and

Gunnarsson, 2006; Hanula et al., 2006). Although previous studies have also shown positive

56 associations between spiders (Araneae) and dead wood (Buddle, 2001; Varady-Szabo and

Buddle, 2006; Hanula et al., 2006), there were no differences in spider abundance, richness or diversity among treatments in this study.

We recently sampled litter-dwelling arthropods near and away from logs in the same forests used in this study (Ulyshen and Hanula, 2009a). Arthropod abundance was significantly higher near logs than away from them as has been shown in the broad-leaved forests of Europe

(Ulyshen and Hanula, 2009a, and references therein). These results are seemingly incongruous with those of the current study. However, it is possible that dead wood affects the distribution of arthropods without affecting their abundance. The pitfall traps used in this study were independently placed with respect to dead wood and the samples were combined from each plot.

Any differences in abundance, richness and diversity due to dead wood proximity were likely canceled out.

It is interesting to note that predators appear to benefit the most from dead wood as compared to other ground-dwelling arthropod taxa. In this study, only ground beetles showed a positive response to the addition of dead wood. Similarly, of the ten arthropod families collected more commonly in reference plots than removal plots by Hanula et al. (2006), four were entirely predatory (Carabidae, Clubionidae, Hahniidae and Lycosidae). Predators may benefit from dead wood if it affects the distribution of their prey, even if it has no effect on prey abundance. It may be easier to locate prey in forests with dead wood given that the abundance of many ground- dwelling arthropod taxa increases with increasing proximity to dead wood (Ulyshen and Hanula,

2009a, and references therein).

2. Saproxylic beetles. There were no differences in total saproxylic beetle abundance, richness or diversity among treatments based on flight intercept trapping or emergence trapping.

57 Furthermore, there were no differences in abundance among treatments for any individual species examined in this study. These results suggest that saproxylic beetles in loblolly pine forests of the southeastern United States have strong dispersal abilities and are little affected by changing amounts of dead wood at the scale of the 9.3 ha plots used in this study. It is important to keep in mind, however, that the plots were embedded in a hospitable forest matrix. Our results may have been quite different had the plots been isolated fragments of forests or surrounded by forests from which all woody material had been removed. It is also important to consider the history of the Savannah River Site because site history is known to strongly influence saproxylic beetle communities. For example, Goßner et al. (2008) found that saproxylic beetles, especially those associated with old wood, were less species rich in forests established on former agricultural land than in forests re-established on ancient woodland sites.

Because the forests used in this study were recently established on former farmland, it is possible that the species most sensitive to manipulations of dead wood were already absent from the forests before the study began. Comparisons between old-growth and second-growth forests are needed to determine whether there are species restricted to old-growth patches in the region before we can fully understand the implications of this research.

3. Management implications. The results from this study are encouraging in that arthropods appear less sensitive than expected to manipulations of dead wood in managed pine forests of the southeastern United States. Because log input plots and removal plots supported equally diverse communities of arthropods, we cannot recommend inputting large amounts of dead wood for conservation purposes, given the expense of such measures. This conclusion is supported by data on shrews, reptiles and amphibians collected by other researchers involved in this study (Moseley et al., 2008; Owens et al., 2008). However, the persistence of saproxylic

58 beetles and other organisms requires that an adequate amount of dead wood is available and we strongly recommend that dead wood be protected whenever possible in managed pine forests. A long history of intensive management in the boreal forests of Finland has resulted in a 90-98% reduction in dead wood at the landscape scale, perhaps threatening more than half all saproxylic species with regional extinction (Siitonen, 2001). If we wish to avoid a similar fate in the southeastern United States, efforts must be made to accommodate dead wood in managed forests. Saproxylic beetles and other arthropods sampled in this study likely owe their apparent resiliency to the quality of the surrounding forest matrix. We support Seymour and Hunter

(1999) in their assertion that the best way to both satisfy timber demands and meet conservation goals is to practice balanced forestry, which they define as “a triad of production forestry and ecological reserves embedded in a matrix of ecological forestry”. This compromise allows for the creation of essential reserves (Niemelä,1997) without decreasing productivity. It also recognizes the importance of the matrix in maintaining diversity at the landscape scale.

Acknowledgements

We thank Scott Horn, Danny Dyer, and Mike Cody for collecting pitfall traps and Walter Sikora,

Ryan Malloy, Stephanie Cahill and Mary Williams for sorting pitfall trap samples. Harry Lee helped identify ground beetles from the pitfall samples and Bob Rabaglia, Chris Carlton and

Alexey Tishechkin assisted in identifying Scolytinae, Pselaphinae and Histeridae, respectively.

We also thank Scott Horn and Mike Cody for helping cut and move the log sections used to sample saproxylic beetles. Support was provided by the Department of Energy-Savannah River

Operations Office through the U.S. Forest Service Savannah River under Interagency Agreement

DE-AI09-00SR22188.

59 Literature Cited

Buddle, C.M., 2001. Spiders (Araneae) associated with downed woody material in a

deciduous forest in central Alberta, Canada. Agricultural and Forest Entomology 3, 241-

251.

Cárcamo, H.A., Parkinson, D., 1999. Distribution of ground beetle assemblages

(Coleoptera, Carabidae) around sour gas processing plants in western Canada.

Pedobiologia 43, 160-173.

Davies, Z.G., Tyler, C., Stewart, G.B., Pullin, A.S., 2008. Are current management

recommendations for saproxylic invertebrates effective? A systematic review.

Biodiversity and Conservation 17, 209-234.

Elton, C.S., 1966. The Pattern of Animal Communities. Methuen and Co Ltd, London.

432 p.

Goßner, M., Engel, K., Jessel, B., 2008. Plant and arthropod communities in young oak stands:

are they determined by site history? Biodiversity and Conservation 17, 3165-3180.

Grove, S.J., 2002. Saproxylic insect ecology and the sustainable management of forests.

Annual Review of Ecology and Systematics 33, 1-23.

Gunnarsson, B., Nittérus, K., Wirdenäs, P. 2004. Effects of logging residue removal on ground-

active beetles in temperate forests. Forest Ecology and Management 201, 229-239.

Hanula, J.L., Horn, S., Wade, D.D., 2006. The role of dead wood in maintaining

arthropod diversity on the forest floor, in: Grove, S.J., Hanula, J.L., (Eds.), Insect

biodiversity and dead wood: proceedings of a symposium for the 22nd International

Congress of Entomology. Gen. Tech. Rep. SRS-93. Asheville, NC: U.S. Department of

Agriculture Forest Service, Southern Research Station. 109 p.

60 Jabin, M., Mohr, D., Kappes, H., Topp, W., 2004. Influence of deadwood on density of

soil macro-arthropods in a managed oak-beech forest. Forest Ecology and Management

194, 61-69.

Jonsell, M., Nittérus, K., Stighäll, K., 2004. Saproxylic beetles in natural and man-made deciduous high stumps retained for conservation. Biological Conservation 118, 163-173.

Kilgo, J.C., Blake, J.I. (Eds.), 2005. Ecology and Management of a Forested Landscape:

Fifty years on the Savannah River Site. Island Press, Washington, DC. 479 p.

Langor, D.W., Hammond, H.E.J., Spence, J.R., Jacobs, J., and Cobb, T.P., 2008. Saproxylic

insect assemblages in Canadian forests: diversity, ecology, and conservation. The

Canadian Entomologist 140, 453-474.

Latty, E.F., Werner, S.M., Mladenoff, D.J., Raffa, K.F., Sickley, T.A., 2006. Response of

ground beetles (Carabidae) assemblages to logging history in northern hardwood-

hemlock forests. Forest Ecology and Management 222, 335-347.

McCay, T.S., Hanula, J.L., Loeb, S.C., Lohr, S.M., McMinn, J.W., Wright-Miley, B.D.,

2002. The role of coarse woody debris in southeastern pine forests: Preliminary results

from a large-scale experiment. USDA Forest Service Gen. Tech. Rep. PSW-GTR-181.

135-144.

Moseley, K.R., Owens, A.K., Castleberry, S.B., Ford, W.M., Kilgo, J.C., McCay, T.S.,

2008. Soricid response to coarse woody debris manipulations in coastal plain loblolly

pine forests. Forest Ecology and Management 255, 2306-2311.

Niemelä, J., 1997. Invertebrates and boreal forest management. Conservation Biology 11,

601-610.

Nittérus, K., Gunnarsson, B., 2006. Effect of microhabitat complexity on the local

distribution of arthropods in clearcuts. Environmental Entomology 35, 1324-1333.

61 Owens, A.K., Moseley, K.R., McCay, T.S., Castleberry, S.B., Kilgo, J.C., Ford, W.M., 2008.

Amphibian and reptile community response to coarse woody debris manipulations in

upland loblolly pine (Pinus taeda) forests. Forest Ecology and Management 256, 2078-

2083.

Pearce, J.L., Venier, L.A., McKee, J., Pedlar, J., McKenney, D., 2003. Influence of

habitat and microhabitat on carabid (Coleoptera: Carabidae) assemblages in four stand

types. Canadian Entomologist 135, 337-357.

Schultz, R.P., 1997. Loblolly pine: The ecology and culture of loblolly pine (Pinus taeda

L.). Agricultural Handbook 713. USDA Forest Service, Washington, DC.

Seymour, R.S., Hunter Jr., M.L., 1999. Principles of ecological forestry, in: Hunter Jr., M.L.

(Ed.), Maintaining Biodiversity in Forest Ecosystems. Cambridge University Press, New

York, pp. 22-61.

Scheckler, S.E., 2001. Afforestation- the first forests, in: Briggs, D.E.G., Crowther, P.R. (Eds.),

Paleobiology II, Blackwell Sciences Ltd, Oxford, pp 67-71.

Siitonen, J., 2001. Forest management, coarse woody debris and saproxylic organisms:

Fennoscandian boreal forests as an example. Ecological Bulletins 49, 11-41.

Ulyshen, M.D., Hanula, J.L., 2007. A comparison of the beetle (Coleoptera) fauna

captured at two heights above the ground in a North American temperate deciduous

forest. American Midland Naturalist 158, 260-278.

Ulyshen, M.D., 2009. Relationships between dead wood and arthropods in the southeastern

United States. Dissertation, University of Georgia.

Ulyshen, M.D., Hanula, J.L., 2009a. Litter-dwelling arthropod abundance peaks near

62 coarse woody debris in loblolly pine forests of the southeastern United States. Florida

Entomologist 92, 163-164.

Ulyshen, M.D., Hanula, J.L., 2009b. Habitat associations of saproxylic beetles in the

southeastern United States: A comparison of forest types, tree species and wood postures.

Forest Ecology and Mangagement 257, 653-664.

Varady-Szabo, H., Buddle, C.M., 2006. On the relationships between ground-dwelling

spider (Araneae) assemblages and dead wood in a northern sugar maple forest.

Biodiversity and Conservation 15, 4119-4141.

63 Table 3.1. Arthropod orders, listed by decreasing abundance, collected in pitfall traps placed in loblolly pine stands on the Savannah River Site, South Carolina.

Order No. morphospecies No. individuals Hymenoptera 239 96 481 Araneae 197 35 012 Coleoptera 407 24 029 Diptera 140 19 064 Julida 1 14 376 Orthoptera 48 9345 Hemiptera 116 3021 Blattaria 4 2946 Opiliones 5 1502 Polydesmida 2 1114 Scolopendromorpha 4 790 Microcoryphia 1 746 Lithobiomorpha 1 589 Chordeumida 3 558 Geophilomorpha 3 269 Pseudoscorpiones 1 219 Psocoptera 5 200 Thysanura 1 128 Lepidoptera 17 101 Thysanoptera 3 92 Spirobolida 1 64 Neuroptera 4 7 Trichoptera 2 2 Mantodea 1 1 Total 1206 210 656

64 Table 3.2. Mean ± SE (n=4) abundance, richness and Shannon’s diversity of arthropods captured in South Carolina, USA. The arthropods were collected with pitfall traps placed in plots with different amounts of coarse woody debris (Reference: no manipulation of dead wood; Log input: 5-fold increase in log volume; Removal: removal of all dead woody material, including snags, ≥ 10 cm in diameter and ≥ 60 cm in length; Snag input: 12-fold increase in snag basal area). Abundance data are presented for orders (and families of Araneae and Coleoptera) represented by ≥ 100 specimens. Richness and diversity data are presented for those taxa represented by ≥ 25 species. For each taxon, means followed by different letters are statistically different (α-0.05) based on Tukey’s

Studentized Range Test. Log(x+1)-transformed abundance data were used for this test, but untransformed data are presented here.

Asterisks denote significance. (Note: Carabidae richness varied significantly among treatments based on ANOVA but there were no differences among means based on Tukey’s Studentized Range Test.)

Order Reference Log input Removal Snag input Family Araneae abundance 2553.0 ± 164.2 2028.3 ± 215.5 2156.5 ± 353.7 2015.3 ± 172.0 richness 86.3 ± 2.7 82.3 ± 3.3 83.0 ± 4.1 86.0 ± 4.1 diversity 3.1 ± 0.1 3.1 ± 0.1 3.2 ± 0.1 3.1 ± 0.0 Agelenidae abundance 161.0 ± 23.5 123.8 ± 19.4 120.3 ± 18.9 122.0 ± 43.1 Anyphaenidae abundance 2.0 ± 1.0 4.0 ± 2.5 76.8 ± 72.5 3.0 ± 1.2 Araneidae abundance 12.8 ± 2.0 5.3 ± 1.8 7.5 ± 1.0 4.5 ± 1.7 Clubionidae abundance 10.3 ± 3.0 15.0 ± 5.5 14.3 ± 3.4 12.0 ± 2.2 Corinnidae abundance 62.5 ± 8.7 40.8 ± 9.5 60.0 ± 10.1 53.0 ± 10.7 Ctenizidae abundance 7.8 ± 2.5 9.5 ± 2.0 10.5 ± 3.1 7.8 ± 1.7 Dictynidae abundance 11.5 ± 4.8 8.5 ± 1.0 12.0 ± 2.2 20.5 ± 3.9 Gnaphosidae abundance 403.5 ± 53.5 398.3 ± 76.4 417.8 ± 98.0 351.0 ± 37.7

65 Hahniidae abundance 324.5 ± 101.0 160.8 ± 93.4 154.8 ± 57.0 160.8 ± 89.0 Linyphiidae abundance 396.8 ± 61.0 311.5 ± 89.8 350.3 ± 37.4 316.8 ± 61.0 richness 16.5 ± 1.0 17.5 ± 1.7 16.8 ± 0.9 17.3 ± 1.3 diversity 2.0 ± 0.1 1.9 ± 0.2 1.9 ± 0.1 1.9 ± 0.1 Lycosidae abundance 972.0 ± 119.8 827.5 ± 113.2 816.0 ± 125.1 834.5 ± 59.8 Pisauridae abundance 9.0 ± 1.8 8.3 ± 1.7 9.0 ± 2.9 9.3 ± 0.8 Salticidae abundance 116.8 ± 46.5 65.5 ± 10.2 54.3 ± 2.9 63.3 ± 19.0 Theridiidae abundance 14.0 ± 1.5 8.5 ± 4.3 9.5 ± 2.1 15.5 ± 2.0 Thomisidae abundance 30.0 ± 11.8 28.0 ± 12.3 24.5 ± 7.5 29.0 ± 7.9 Blattaria abundance 188.5 ± 20.1 175.3 ± 28.8 181.0 ± 59.1 191.8 ± 33.0 Chordeumida abundance 32.0 ± 8.6 41.0 ± 8.9 36.5 ± 20.3 30.0 ± 9.7 Coleoptera abundance 1654.8 ± 144.9 1385.3 ± 186.9 1488.0 ± 146.6 1479.3 ± 170.3 richness 138.0 ± 7.6 141.0 ± 6.1 136.0 ± 3.4 131.3 ± 3.1 diversity 3.6 ± 0.1 3.8 ± 0.1 3.6 ± 0.1 3.6 ± 0.1 Carabidae abundance 324.0 ± 62.1 286.5 ± 53.1 291.0 ± 46.3 303.3 ± 44.6 richness* 22.3 ± 1.4 26.8 ± 1.3 22.0 ± 1.4 22.3 ± 0.3 diversity* 1.7 ± 0.1 b 2.3 ± 0.1 a 1.7 ± 0.1 b 1.8 ± 0.1 ab Chrysomelidae abundance 19.8 ± 5.4 16.5 ± 2.8 10.0 ± 2.4 14.8 ± 6.6 Ciidae abundance 9.8 ± 2.3 11.0 ± 3.1 5.3 ± 2.3 10.0 ± 3.3 Cryptophagidae abundance 44.8 ± 16.6 18.8 ± 4.1 33.5 ± 2.9 31.0 ± 1.8 Curculionidae abundance 182.5 ± 18.6 184.5 ± 13.6 168.3 ± 38.4 188.3 ± 9.7 richness 14.5 ± 1.3 14.8 ± 1.1 13.5 ± 1.7 13.0 ± 1.9 diversity 1.7 ± 0.1 1.6 ± 0.1 1.7 ± 0.2 1.5 ± 0.1 Elateridae abundance 13.3 ± 1.8 10.8 ± 2.9 19.3 ± 2.7 15.8 ± 3.1 Endomychidae abundance 8.5 ± 1.9 9.8 ± 2.0 6.0 ± 1.9 3.5 ± 1.0 Erotylidae abundance 46.5 ± 20.6 26.3 ± 8.1 95.3 ± 54.7 68.5 ± 32.5 Histeridae abundance 7.0 ± 3.2 5.0 ± 1.1 10.3 ± 4.6 7.0 ± 2.5 Leiodidae abundance 40.8 ± 9.6 28.0 ± 2.7 33.3 ± 8.5 26.5 ± 1.4 Melyridae abundance 10.3 ± 9.3 5.5 ± 3.1 8.8 ± 5.7 1.8 ± 1.8 Nitidulidae abundance 40.0 ± 13.3 34.0 ± 12.3 26.8 ± 6.4 18.0 ± 5.1 Scarabaeidae abundance 93.3 ± 15.5 97.8 ± 12.0 105.3 ± 16.8 119.5 ± 40.7 richness 18.0 ± 1.1 18.3 ± 1.3 17.5 ± 0.6 18.3 ± 2.2 diversity 2.3 ± 0.2 2.4 ± 0.1 2.3 ± 0.1 2.1 ± 0.2

66 Scydmaenidae abundance* 116.0 ± 8.4 a 63.3 ± 16.8 ab 66.8 ± 18.8 ab 53.0 ± 13.7 b Staphylinidae abundance 595.5 ± 129.1 486.5 ± 130.7 518.5 ± 162.5 511.0 ± 114.0 richness 22.8 ± 1.2 19.5 ± 2.7 19.8 ± 1.0 20.5 ± 1.3 diversity 1.8 ± 0.1 1.9 ± 0.0 2.0 ± 0.1 1.9 ± 0.1 Tenebrionidae abundance 60.0 ± 11.7 56.8 ± 5.7 49.5 ± 3.8 73.8 ± 33.1 Diptera abundance 1297.3 ± 235.0 1297.3 ± 296.5 966.5 ± 143.1 1205.0 ± 228.8 richness 53.0 ± 1.6 51.3 ± 1.4 49.0 ± 2.7 47.8 ± 1.9 diversity 2.4 ± 0.1 2.3 ± 0.1 2.4 ± 0.1 2.2 ± 0.1 Geophilomorpha abundance 16.8 ± 2.0 12.3 ± 0.5 22.3 ± 2.2 16.0 ± 4.1 Hemiptera abundance 188.8 ± 25.6 242.0 ± 64.4 131.3 ± 24.7 193.3 ± 24.9 richness 33.8 ± 1.8 35.8 ± 4.3 32.0 ± 4.3 32.5 ± 0.6 diversity 2.7 ± 0.1 2.5 ± 0.1 2.7 ± 0.1 2.4 ± 0.1 Hymenoptera abundance 5422.5 ± 174.4 5006.0 ± 997.8 7004.8 ± 912.3 6687.0 ± 1485.3 richness 88.0 ± 2.5 79.8 ± 4.8 80.3 ± 3.1 82.8 ± 1.7 diversity 2.5 ± 0.0 2.2 ± 0.2 2.2 ± 0.1 2.1 ± 0.2 Julida abundance 1055.8 ± 282.3 744.8 ± 180.0 1162.3 ± 299.5 631.3 ± 123.0 Lepidoptera abundance 5.8 ± 1.5 6.5 ± 2.9 5.5 ± 1.2 7.5 ± 1.5 Microcoryphia abundance 48.5 ± 29.4 69.8 ± 40.4 48.8 ± 32.1 19.5 ± 16.0 Opiliones abundance 110.5 ± 25.5 105.8 ± 33.5 120.3 ± 14.8 39.0 ± 6.0 Orthoptera abundance* 666.5 ± 55.3 a 663.3 ± 83.7 a 598.0 ± 127.7 ab 408.5 ± 64.2 b richness 18.8 ± 1.4 19.0 ± 1.1 20.0 ± 0.4 16.8 ± 0.9 diversity 1.7 ± 0.0 1.8 ± 0.1 1.9 ± 0.1 1.9 ± 0.1 Polydesmida abundance* 44.0 ± 24.1 ab 96.5 ± 58.2 a 86.0 ± 65.4 ab 52.0 ± 43.4 b Pseudoscorpiones abundance 10.5 ± 4.6 13.8 ± 2.0 12.5 ± 3.7 18.0 ± 6.2 Psocoptera abundance 13.3 ± 4.4 9.5 ± 4.1 8.5 ± 0.9 18.8 ± 10.0 Scolopendromorpha abundance 42.0 ± 5.3 49.5 ± 10.0 60.0 ± 19.0 46.0 ± 8.6 Thysanura abundance 4.8 ± 1.8 11.3 ± 4.5 9.8 ± 6.5 6.3 ± 2.1 TOTAL abundance 13 405.0 ± 637.3 12 006.8 ± 1384.0 14 143.0 ± 1537.8 13 109.3 ± 1481.2 richness 440.5 ± 11.2 434.3 ± 16.5 425.8 ± 7.8 421.5 ± 6.6 diversity 4.2 ± 0.0 4.1 ± 0.1 3.9 ± 0.1 3.8 ± 0.2

67 Table 3.3. Mean ± SE (n=4) richness and diversity of Coleoptera collected in flight intercept traps placed in plots with different amounts of coarse woody debris (Reference: no manipulation of dead wood; Log input: 5-fold increase in log volume; Removal: removal of all dead woody material, including snags, ≥ 10 cm in diameter and ≥ 60 cm in length).

Reference Log input Removal All species richness 76.3 ± 3.5 76.5 ± 3.1 80.3 ± 2.7 diversity 3.0 ± 0.2 2.9 ± 0.2 2.8 ± 0.2 Saproxylic species only richness 23.5 ± 1.2 26.0 ± 3.1 22.0 ± 1.6 diversity 2.2 ± 0.1 2.4 ± 0.1 2.2 ± 0.2

68 Table 3.4. ANOVA table for a split-split plot design with saproxylic beetle species richness as the response variable. This randomized complete block design consisted of three treatments (i.e., whole plots: log removal, log input and reference). Sampling took place at three locations (i.e., split plots: edge, center, and half-way between edge and center) within each whole plot and at four different times (i.e., split-split plots: 2, 6, 10 and 22 months) at each location. The asterisk denotes significance (P<0.05).

Source df MS F Whole plots treatment 2 25.76 F2,6=0.63 block 3 10.04 F3,6=0.24 treatment*block 6 40.99 - Split plots location 2 12.72 F2,18=0.30 treatment*location 4 19.69 F4,18=0.47 block *location 6 73.36 - block *treatment*location 12 25.91 - error (location) (18) 41.73 - Split-split plots date 3 330.71 F3,81=16.51 * treatment*date 6 5.02 F6,81=0.25 location*date 6 17.03 F6,81=0.85 treatment*location*date 12 28.61 F12,81=1.43 block *date 9 19.90 - block *treatment*date 18 21.64 - block *location*date 18 22.18 - block *treatment*location*date 36 18.18 - error (date) (81) 20.03 - Total 143

69 Table 3.5. ANOVA results (F-values) for the most common saproxylic beetle species to emerge from loblolly pine logs. Sampling followed a split-split plot design and abundance was the response variable in the analyses. This randomized complete block design consisted of three treatments (i.e., whole plots: log removal, log input and reference). Sampling took place at three locations (i.e., split plots: edge, center, and half-way between edge and center) within each whole plot and at four different times (i.e., split-split plots: 2,

6, 10 and 22 months) at each location. The asterisk denotes significance (P<0.05).

Species treatment location date treat*loc treat*date loc*date treat*loc*date Family F2,6 F2,18 F3,81 F4,18 F6,81 F6,81 F12,81 Aderidae Cnopus impressus (LeConte) 1.04 2.19 8.81 * 2.16 0.88 0.64 2.21 * Ganascus ventricosus (LeConte) 0.79 1.22 14.62 * 2.60 0.16 0.92 1.16 Biphyllidae Diplocoelus rudis (LeConte) 1.18 0.61 5.51 * 0.62 0.82 1.20 1.19 Bostrichidae Stephanopachys sp. 4.41 1.58 5.93 * 4.42 1.04 0.53 0.72 Buprestidae Buprestis lineata Fabricius 0.13 1.84 59.76 * 0.31 0.14 1.76 0.38 Chalcophora virginiensis Drury 0.35 1.25 8.81 * 2.12 0.48 0.99 1.52 Carabidae Mioptachys flavicauda (Say) 0.75 1.24 3.02 * 0.93 0.62 1.15 0.52 Tachyta nana inornata (Say) 1.65 1.82 3.29 * 2.81 2.10 0.84 1.50 Cerambycidae Acanthocinus obsoletus (Olivier) 1.59 1.78 81.11 * 1.56 4.01 * 2.25 * 1.15 Monochamus titillator (Fabricius) 0.02 0.48 29.94 * 0.32 0.20 0.19 0.56 Cerylonidae Philothermus glabriculus LeConte 1.81 0.31 24.48 * 2.49 1.30 0.26 2.24 * Ciidae Ciidae spp. 0.38 0.61 34.47 * 1.50 0.25 0.93 1.77 Curculionidae Cossonus spp. 0.82 0.66 15.80 * 0.60 0.56 0.92 0.74 Elateridae Lacon impressicollis (Say) 0.51 0.40 10.88 * 0.68 0.68 0.37 0.71 Histeridae Bacanius punctiformis (LeConte) 2.36 0.03 5.25 * 1.42 0.64 0.55 0.79 Mycetophagidae Mycetophagus pini Ziegler 0.09 0.22 31.25 * 0.79 0.14 0.19 0.70 Ptiliidae Ptiliidae spp. 0.80 0.49 12.20 * 1.23 0.76 0.51 1.29 Staphylinidae Anacyptus testaceus (LeConte) 0.34 0.78 1.66 1.60 1.79 0.72 0.70 Sepedophilus sp. 3 2.05 2.40 3.38 * 0.88 1.26 1.01 1.35 Staphylinidae sp. 65 1.02 0.40 4.51 * 0.41 1.64 0.24 1.03 Tenebrionidae Hymenorus sp. 1.16 0.28 5.26 * 0.34 1.61 0.26 0.27 Platydema flavipes (Fabricius) 0.04 0.43 20.83 * 0.80 0.59 0.74 2.01 Platydema ruficorne (Stürm) 3.01 0.46 3.92 * 0.55 0.81 0.74 0.43 Zopheridae Pycnomerus haematodes (Fabricius) 1.10 1.25 8.00 * 1.26 0.74 0.96 1.16 Pycnomerus sulcicollis LeConte 1.78 0.60 4.64 * 0.77 1.22 3.27 * 0.49

70 Table 3.6. Ground beetles collected in pitfall traps placed in plots with different amounts of coarse woody debris in the southeastern United States (Reference: no manipulation of dead wood; Log input: 5-fold increase in log volume; Removal: removal of all dead woody material, including snags, ≥ 10 cm in diameter and ≥ 60 cm in length; Snag input: 12-fold increase in snag basal area).

Log Snag Reference input Removal input Total Acupalpus spp. 2 11 2 4 19 Agonum punctiforme (Say) 2 1 1 1 5 Amara cupreolata Pultzeys 70 44 49 66 229 Amara sp. 5 4 2 3 14 Amblygnathus iripennis (Say) 2 0 0 0 2 Anisodactylus merula (Germar) 17 34 7 12 70 Apenes sinuatus (Say) 2 6 3 4 15 Bembidiini sp. 1 4 9 2 2 17 Bembidiini sp. 2 1 0 0 2 3 Bembidiini sp. 3 1 0 0 0 1 Bembidiini sp. 4 2 1 1 1 5 Brachinus alternans Dejean 0 0 1 1 2 Calathus opaculus LeConte 32 29 10 21 92 Calosoma calidum (Fabricius) 0 1 0 0 1 Carabus sylvosus Say 50 77 72 48 247 Chlaenius amoenus Dejean 3 15 5 6 29 Chlaenius tomentosus (Say) 0 0 0 1 1 Cicindela abdominalis Fabricius 0 0 2 0 2 Cicindela sexguttata Fabricius 0 2 0 0 2 Cicindela unipunctata Fabricius 1 1 0 0 2 Clivina sulcipennis Putzeys 0 1 0 4 5 Cyclotrachelus convivus (LeConte) 41 57 8 53 159 Cyclotrachelus laevipennis (LeConte) 6 21 61 11 99 Cyclotrachelus levifaber (Freitag) 465 270 281 334 1350 Cymindis americanus Dejean 11 15 15 9 50 Cymindis limbatus Dejean 0 4 3 3 10 Dicaelus dilatatus Say 8 17 11 10 46 Dicaelus furvus Dejean 29 41 22 22 114 Dyschiriodes sphaericollis (Say) 16 10 7 8 41 Galerita bicolor (Drury) 0 1 0 0 1 Harpalus katiae Battoni 0 1 0 0 1 Harpalus protractus Casey 17 14 18 28 77

71 Helluomorphoides clairvillei (Dejean) 13 6 15 29 63 Helluomorphoides nigripennis (Dejean) 3 8 7 9 27 Loxandrus crenatus LeConte 1 2 0 6 9 Loxandrus sp. 1 0 0 0 1 Megacelphala virginica (Linnaeus) 2 4 2 2 10 Megacephala carolina (Linnaeus) 0 2 0 0 2 Notiobia terminata (Say) 1 0 0 0 1 Notiophilus spp. 438 312 501 438 1689 Pasimachus marginatus (Fabricius) 0 6 1 4 11 Pasimachus sublaevis (Palisot de Beauvois) 1 0 0 0 1 Pasimachus subsulcatus Say 8 44 12 31 95 Pentagonica flavipes (LeConte) 0 1 0 1 2 Pentagonica picticornis Bates 0 1 1 0 2 Piesmus submarginatus (Say) 2 2 2 7 13 Poecilus lucublandus (Say) 0 1 0 0 1 Pterostichus sculptus LeConte 0 39 5 2 46 Scarites subterraneus Fabricius 0 0 0 1 1 Selenophorus opalinus (LeConte) 14 10 4 2 30 Selenophorus spp. 19 12 24 22 77 Stenolophus plebejus Dejean 5 7 6 5 23 Tetragonoderus intersectus (Germar) 0 2 1 0 3 Zuphium americanum Dejean 1 0 0 0 1 Total (individuals/species) 1296/37 1146/43 1164/35 1213/38 4819/54

72 Table 3.7. List of saproxylic beetles known from loblolly pine captured in flight intercept traps.

Family Species Number Anthribidae Euparius marmoreus (Olivier) 7 Buprestidae Chalcophora virginiensis Drury 1 Carabidae Mioptachys flavicauda Say 3 Cerambycidae Eupogonius tomentosus (Haldeman) 1 Typocerus zebra (Olivier) 545 Cerylonidae Philothermus glabriculus LeConte 3 Ciidae Ciidae spp. 9 Colydiidae Bitoma quadriguttata (Say) 1 Namunaria guttula (LeConte) 1 Synchita fuliginosa Melsheimer 2 Corylophidae Corylophidae spp. 59 Curculionidae Dryophthorus americanus Bedel 5 Elateridae Ampedus areolatus (Say) 2 Lacon discoidea (Weber) 1 Lacon impressicollis (Say) 2 Melanotus ignobilis Melsheimer 29 Melanotus sp. 2 19 Endomychidae Mycetina perpulchra (Newman) 1 Eucnemidae Dromaeolus sp. 8 Fornax sp. 19 Histeridae Bacanius punctiformis (LeConte) 7 Laemophloeidae Lathropus vernalis LeConte 1 Leiodidae sp. 1 11 sp. 2 14 Lycidae Plateros sp. 20 Melyridae Attalus scincetus (Say) 2 Monotomidae Bactridium sp. 11 Monotoma sp. 2 Mycetophagidae Litargus tetraspilotus LeConte 4 Mycetophagus pini Ziegler 1 Passandridae Catogenus rufus (Fabricius) 2 Ptiliidae Ptiliidae spp. 2 Scraptiidae Canifa sp. 22 Silvanidae Ahasversus advena (Waltl) 1 Ahasversus rectus (LeConte) 13 Cathartosilvanus imbellis (LeConte) 2 Silvanus muticus Sharp 1 Sphindidae Sphindus sp. 41 Staphylinidae Anacyptus testaceus (LeConte) 1 Clavilispinus sp. 4 Lordithon obsoletus (Say) 3

73 Myrmecocephalus sp. 25 Palaminus sp. 8 Scaphidiinae sp. 2 2 Scaphidiinae sp. 3 4 Sepedophilus sp. 10 Thoracophorus costalis (Erichson) 38 Staphylinidae sp. 75 1 Staphylinidae sp. 80 3 Tenebrionidae Hymenorus sp. 258 Lobopoda erythrocnemis Germar 249 Platydema flavipes (Fabricius) 10 Platydema ruficorne (Stürm) 5 Uloma punctulata LeConte 1 Tetratomidae Eustrophopsis bicolor (Fabricius) 75 Holostrophus bifasciatus (Say) 1 Throscidae Throscidae spp. 94 Zopheridae Hyporhagus punctulatus Thomson 54 Pycnomerus haematodes (Fabricius) 13 Pycnomerus sulcicollis LeConte 35 Total 1769

74 Table 3.8. Saproxylic beetles species that emerged from loblolly pine logs.

Family Species Number Aderidae Cnopus impressus (LeConte) 825 Ganascus ventricosus (LeConte) 143 Pseudariotus notatus (LeConte) 6 Aderidae sp. 4 1 Aderidae sp. 5 5 Anobiidae Anobiidae sp. 1 1 Anthribidae Euparius marmoreus (Olivier) 60 Euparius paganus Gyllenhal 33 Biphyllidae Diplocoelus rudis (LeConte) 146 Bostrichidae Stephanopachys sp. 166 Bruchidae Bruchidae sp. 1 Buprestidae Buprestis lineata Fabricius 657 Buprestis maculipennis Gory 2 Chalcophora virginiensis Drury 150 Chrysobothris dentipes Germar 7 Chrysobothris floricola Gory 6 Dicerca punctulata (Schönherr) 2 Carabidae Mioptachys flavicauda (Say) 257 Piesmus submarginatus (Say) 5 Tachyta nana inornata (Say) 141 Carabidae sp. 1 Cerambycidae Acanthocinus nodosus (Fabricius) 3 Acanthocinus obsoletus (Olivier) 194 Asemum striatum (Linnaeus) 1 Astylopsis sexguttata (Say) 12 Eupogonius tomentosus (Haldeman) 2 Monochamus titillator (Fabricius) 129 Typocerus zebra (Olivier) 77 Xylotrechus sagittatus (Germar) 89 Cerylonidae Cerylon unicolor (Ziegler) 7 Philothermus glabriculus LeConte 107 Chrysomelidae Cryptocephalinae sp. 1 3 Cryptocephalinae sp. 2 1 Ciidae Ciidae spp. 1917 Cleridae Priocera castanea (Newman) 7 Colydiidae Bitoma quadriguttata (Say) 93 Colydium nigripenne LeConte 40 Endeitoma dentata (Horn) 23 Endeitoma granulata (Say) 45 Lasconotus sp. 1 Namunaria guttulata (LeConte) 69

75 Paha laticollis (LeConte) 8 Corylophidae Corylophidae spp. 61 Cryptophagidae Cryptophagidae sp. 4 3 Curculionidae Carphoborus bifurcus Eichhoff 5 Cossonus spp. 4635 Crypturgus alutaceus Schwarz 6 Dryophthorus americanus Germar 39 Gnathotrichus materiarius (Fitch) 1 Himatium errans LeConte 2 Hylastes porculus Erichson 1 Hylastes salebrosus Eichhoff 4 Hylastes tenuis Eichhoff 1 Hypothenemus sp. 6 Ips avulsus (Eichhoff) 2 Ips grandicollis (Eichhoff) 62 Myoplatypus flavicornis (Fabricius) 1 Orthotomicus caelatus (Eichhoff) 30 Pachylobius picivorus (Germar) 6 Pityophthorus sp. 1 Scolytinae sp. 1 Xyleborus pubescens Zimmermann 6 Curculionidae sp. 30 1 Dermestidae Dermestidae sp. 1 Elateridae Alaus myops (Fabricius) 7 Dicrepidius ramicornis (Palisot de Beauvois) 7 Dipropus spp. 38 Drapetes geminatus Say 5 Drapetes quadripustulatus Bonvouloir 3 Lacon discoidea (Weber) 5 Lacon impressicollis (Say) 102 Melanotus ignobilis Melsheimer 7 Melanotus sp. 2 3 Endomychidae Aphorista vittata (Fabricius) 10 Endomychus biguttatus Say 3 Mycetina perpulchra (Newman) 19 Eucnemidae Dromaeolus sp. 52 Histeridae Aeletes simplex (LeConte) 7 Bacanius punctiformis (LeConte) 195 Bacanius sp. 2 3 Eblisia carolina (Paykull) 2 Epierus pulicarius Erichson 13 Paromalus seminulum Erichson 1 Platysoma coarctatum LeConte 4 Platysoma cylindricum (Paykull) 14 Platysoma parallelum (Say) 25

76 Plegaderus transversus (Say) 6 Laemophloeidae Cryptolestes punctatus (LeConte) 4 Cryptolestes sp. 1 Melyridae Attalus circumscriptus (Say) 1 Attalus otiosus (Say) 8 Attalus scincetus (Say) 7 Melyridae sp. 5 2 Melyridae sp. 6 2 Micromalthidae Micromalthus debilis LeConte 1 Monotomidae Monotoma sp. 1 Mordellidae Mordellidae sp. 3 5 Mordellidae sp. 4 3 Mycetophagidae Mycetophagus obsoletus (Melsheimer) 1 Mycetophagus pini Ziegler 944 Nitidulidae Nitidulidae sp. 10 1 Oedemeridae Oxycopis thoracica (Fabricius) 2 Passandridae Catogenus rufus (Fabricius) 11 Phalacridae Phalacridae sp. 3 2 Ptiliidae Ptiliidae spp. 416 Scarabaeidae Scarabaeidae sp. 2 1 Valgus seticollis (Palisot de Beauvois) 17 Scraptiidae Canifa sp. 50 Scydmaenidae Scydmaenidae spp. 11 Silvanidae Ahasversus advena (Waltl) 1 Ahasversus rectus (LeConte) 5 Cathartosilvanus imbellis (LeConte) 1 Silvanus muticus Sharp 36 Sphindidae Sphindus sp. 2 Staphylinidae Anacyptus testaceus (LeConte) 161 Clavilispinus sp. 56 Coproporus ventriculus (Say) 1 Dalmosanus sp. 3 Gyrohypnus sp. 9 Myrmecocephalus sp. 1 Scaphidiinae sp. 2 6 Sepedophilus sp. 3 116 Staphylinidae sp. 29 1 Staphylinidae sp. 65 582 Staphylinidae sp. 67 36 Staphylinidae sp. 68 19 Staphylinidae sp. 75 4 Staphylinidae sp. 76 4 Staphylinidae sp. 77 1 Staphylinidae sp. 78 3 Staphylinidae sp. 80 50 Staphylinidae sp. 82 4

77 Staphylinidae sp. 83 1 Staphylinidae sp. 84 34 Thoracophorus costalis (Erichson) 81 Tenebrionidae Corticeus parallelus (Melsheimer) 1 Corticeus thoracicus (Melsheimer) 2 Hymenorus sp. 283 Platydema flavipes (Fabricius) 672 Platydema ruficorne (Stürm) 840 Statira sp. 1 Tharsus seditiosus LeConte 64 Uloma imberbis LeConte 20 Uloma punctulata LeConte 11 Tetratomidae Eustrophopsis bicolor (Fabricius) 4 Holostrophus bifasciatus (Say) 3 Throscidae Aulonothroscus convergens (Horn) 15 Aulonothroscus sp. 57 Trogossitidae Temnoscheila virescens (Fabricius) 31 Tenebroides collaris (Sturm) 2 Tenebroides corticalis (Melsheimer) 28 Tenebroides marginatus (Palisot de Beauvois) 13 Tenebroides obtusus (Horn) 2 Zopheridae Pycnomerus haematodes (Fabricius) 229 Pycnomerus sulcicollis LeConte 511 Total 16 347

78 Fig. 3.1. Research plots (grouped by block) on the Savannah River Site, South Carolina, USA.

Ground-dwelling arthropods were sampled in all plots using pitfall traps. To sample saproxylic beetles, groups of logs were placed on the edge, at the center, and halfway between the edge and center of each log input, log removal and reference plot. Approximate placement is indicated by the lines half-bisecting each of those plots. Flight intercept traps were placed near the center of each log input, log removal and reference plot. Note: snag input plots were only sampled with pitfall traps.

79

80 Fig. 3.2. Circular arrangement of logs used to sample saproxylic beetles at three locations in each log input, log removal and reference plot. Saproxylic beetles were reared from logs removed randomly from each location after 2, 6, 10 and 22 months.

81 82 CHAPTER 4

PATTERNS OF SAPROXYLIC BEETLE SUCCESSION IN LOBLOLLY PINE¹

¹Ulyshen, M.D. and J.L. Hanula. Submitted to Canadian Journal of Forest Research,

10/12/2008

83 Abstract

Patterns of insect succession in dead wood remain unclear, particularly beyond the first several years of decay. Because beetle species inhabiting highly decayed wood are often the most vulnerable in European forests, it is important to become better acquainted with their poorly known counterparts in North America, particularly in commercially important and intensively managed tree species such as loblolly pine. In this study, saproxylic beetles were sampled from loblolly pine logs at three stages of decay (1-15, 46-60 and 94-108 months after the trees were killed) using 1) emergence traps attached to logs in the field and 2) rearing bags in the laboratory. Species richness peaked within the first year, due to a diverse assemblage of phloem- feeders and predators associated with young logs. Estimated species richness also peaked within the first year and declined steadily thereafter. The three decay classes formed distinct groupings based on nonmetric multidimensional scaling and 25 and 7 species were significantly associated with young and old logs, respectively, based on indicator species analysis. Because species composition differed among decay classes and a number of species were significantly associated with the oldest logs sampled in this study, it is important to protect wood of all age classes in managed loblolly pine forests.

84 Introduction

In contrast to rapidly degrading substrates like carrion and dung, wood decays slowly, often requiring decades or even centuries to decompose completely, and cannot be studied easily from beginning to end. This process is accelerated, perhaps by as much as 70% (Dajoz 2000, and reference therein), by the activities of saproxylic (i.e., dependent on dead wood) insects. The most conspicuous and species rich of these are beetles. Beetle community composition changes during wood decay in a predictable fashion. Although the details of this succession remain poorly understood, particularly at advanced stages of decay, the process can be partitioned into three overlapping phases (e.g., Savely 1939): 1) Phloem phase: The first beetles to colonize dead wood are phloem feeders (e.g., scolytine curculionids, cerambycids, buprestids, etc.) and their predators. These beetles consume the carbohydrate-rich cambium layer, eventually (i.e., usually within weeks) causing the bark to separate from the wood. Because phloeophagous beetles must cope with the chemical defenses of recently-killed trees, they are often highly host specific; 2) Subcortical space phase: The subcortical space resulting from the activities of the phloem-feeding beetles is rapidly colonized by fungus and a diverse assemblage of beetles (e.g., laemophloeids, silvanids, nitidulids, etc.) specialized for life under bark. This phase ends within months or years when the bark finally falls away from the wood. Because subcortical beetles are generally mycophagous or predatory and do not feed on the wood itself, they are generally less tree-species specific than phloem feeders (Elton 1966); 3) Rot phase: As decomposition proceeds, wood becomes increasingly infiltrated by fungi. The beetle communities consequently become increasingly dominated by mycophages and their predators. The beetles inhabiting wood at these more advanced stages are the least specific to particular tree species (Howden and

Vogt 1951). How insect species richness changes during the decay process is poorly understood.

85 It may either increase with more advanced decay classes in response to the diversification of wood-rotting fungi and increased microhabitat diversity (Langor et al. 2008, and references therein) or decrease in response to declining nutritional quality following the loss of phloem

(Howden and Vogt 1951; Siitonen 2001). This basic question remains largely unresolved because few efforts have been made to sample insects from a sufficiently wide range of age classes under carefully controlled experimental conditions.

Because a number of bark beetles attack weakened trees and are considered important pests, the beetle communities associated with recently killed trees have been studied in greater detail (e.g., Overgaard 1968) than those associated with later stages of decay. The total amount of time required for dead wood to decompose completely differs among forest types and climates with factors such as temperature and humidity being particularly important (Elton 1966). The following negative exponential model has been widely used to describe the decay rate of dead

-kt wood: Yt=Y0e , where Yt is the amount of material at time t, Y0 is the initial amount of material, and k is the decay rate constant (Harmon et al. 1986). The decay rate constant varies with climate and among tree species. Barber and Van Lear (1984) determined that the overall decay rate constant for loblolly pine (Pinus taeda L.) slash in the Piedmont of South Carolina was

0.072. Based on this rate, 50% of loblolly pine slash should be gone after 10 yr, 90% after 32 yr, and 99% after 64 yr (Barber and Van Lear 1984). Because the cambium layer is degraded within months and bark has fallen away within several years, beetle succession in loblolly pine is only well understood for the first few years (i.e., the phloem and subcortical space phases) of this >64 year process.

Interest in saproxylic beetle conservation has increased recently in response to evidence that many species are at risk of disappearing in intensively managed forests. Saproxylic beetles

86 with limited dispersal abilities are particularly at risk, especially within increasingly fragmented landscapes (Jonsson 2003; Scheigg 2000). Dispersal ability varies greatly among saproxylic insects and species dependent on ephemeral resources are thought to be better dispersers than those found in more stable habitats (Nilsson and Baranowski 1997). Because the nature of dead wood at later stages of decay is more permanent than that of earlier stages, it may be reasonable to suppose that the first species to colonize dead trees (i.e., phloem feeders) are better dispersers than those colonizing trees at later stages of decay. If so, it follows that early colonizers should be less negatively affected by forest management and fragmentation. Indeed, researchers in

Scandinavia have shown that bark beetles (i.e., early colonists) can fly long distances in order to colonize freshly killed wood (Nilssen 1984) and are little affected by forest management

(Johansson et al. 2006). There is also evidence from Tasmania that beetles inhabiting stable dead wood habitats (i.e. brown rot in large-diameter logs) have relatively poor dispersal powers

(Yee et al. 2006) and most endangered species in Europe are associated with wood in advanced stages of decay (Langor et al. 2008, and references therein). Because beetle species inhabiting wood at later stages of decay may be among the most vulnerable, it is important to become better acquainted with this still poorly known community, particularly in commercially important (i.e., intensively managed) tree species such as loblolly pine.

Loblolly pine has largely displaced longleaf pine (P. palustris Mill.) throughout the southeastern United States and is currently one of the most intensively managed timber trees in

North America (Schultz 1997). Yet the saproxylic beetle communities associated with loblolly pine and other species of pine native to the eastern United States remain poorly understood.

Although the beetle species present within the first year have received considerable attention

(Graham 1925; Overgaard 1968; Moser et al. 1971; Hines and Heikkenen 1977), detailed studies

87 at later stages of decay have been rare. Savely (1939) studied the succession of beetles in

Virginia pine (Pinus virginiana Mill.) logs at various stages of decay, including some that were more than three years old, in North Carolina. Howden and Vogt (1951) collected beetles from

Virginia pine snags in Maryland including some broken off trunks that may have been more than ten years old. How closely patterns of beetle succession in Virginia pine resemble those in loblolly pine is not known and, to our knowledge, there have been no attempts to study the beetle communities present in loblolly pine at advanced stages of decay. This study was designed to answer the following questions regarding the first ten years of beetle succession in loblolly pine:

1) Which stage of decay supports the most species rich beetle communities and how does species richness change during the decay process? 2) How does community composition change during the decay process? 3) Which beetle species are associated with highly decayed wood?

Materials and Methods

Study Site

This research took place on the 80,267-ha Savannah River Site (SRS) located in the upper Coastal Plain Physiographic Province of South Carolina. The SRS, a facility owned and operated by the United States Department of Energy, was established in 1951, and was designated an Environmental Research Park in 1972 (Kilgo and Blake 2005). Most of the land now owned by the Savannah River site was formerly used for agricultural purposes and most forests currently standing, including those used in this study, were planted in the early 1950’s

(Kilgo and Blake 2005). Sampling took place in and around long-term research plots (see

Ulyshen and Hanula, in review) established on the SRS to explore relationships between coarse woody debris and animal communities in loblolly pine (Pinus taeda L.) forests. Loblolly pine

88 was the dominant tree species but several other species, including water oak (Quercus nigra L.) and sweetgum (Liquidambar styraciflua L.), were also present at low densities. The understory was generally dominated by wax myrtle (Myrica cerifera L.), blackberry (Rubus spp.), kudzu

(Pueraria montana (Lour.) Merr.), Lespedeza bicolor Turcz., and Japanese honeysuckle

(Lonicera japonica Thunb.).

Insect Sampling

This was a two-part project, using two different methods to collect insects emerging from loblolly pine (P. taeda L.) logs. The first method, “on-location sampling,” involved trapping insects directly from logs in the field. The second method, “off-location sampling,” involved removing sections from the logs and transporting them to an emergence facility located in

Athens, Georgia. The trees used in this study were planted in the early 1950’s and were felled on the following dates: August 1997 (“old” logs), August 2001 (“middle-aged” logs) and May 2005

(“young” logs). The young logs were completely covered in bark at the beginning of the study whereas the middle-aged and old logs were completely bark-free.

1. On-location sampling. On-location sampling followed a completely randomized block design with four blocks (see map in Ulyshen and Hanula, in review). Emergence traps (Fig. 4C) were installed on four logs of each age class in each block (i.e., 12 logs per block and 48 traps in total). Each emergence trap consisted of black cotton canvas wrapped around a log and attached to a clear plastic PVC pipe (length: 101 cm, OD: 11.4 cm, ID: 10.2 cm, Excalibur Extrusions

Inc., Placentia, CA) (Fig. 4C). Emerging beetles entered the pipe through an opening (84 x 6 cm) facing the log. The section of pipe removed to create this opening was attached above the opening with three wing-nuts (Fig. 4C). One end of the pipe was capped. The other end led, via

89 a PVC elbow connector, to a collecting jar filled with propylene glycol (Fig. 4C). The pipe was supported by positioning the opening over a pair of long nails driven at angles into one side of the log. Additional nails were driven into the top and opposite side of the log to create a space between the log and the cloth. The cloth was fastened to the pipe below the opening with closely-spaced screws. After wrapping the cloth around the log, the loose end of cloth was pinched between the pipe and the piece of pipe removed to create the opening using three wing- nuts (Fig. 4C). The cloth was then tightly bound to the log on each side of the pipe with metal wire. The average length of wood enclosed within the traps, as measured by the distance between the wires, was 1.12 m. When logs were in contact with the soil, shovels were used to create a space through which to pass the cloth. Some of the oldest logs were too decayed to support the traps using nails. In such cases, sections of wood were removed and enclosed within traps by tying off the two ends of cloth. We sampled continuously from 7 June 2005 to 16

March 2006.

2. Off-location sampling. Sections were removed from 12 logs, four from each of the three age classes, on August 8 2006. Chainsaws were used to remove three 0.5 m sections from each log for a total of 36 sections. The three sections taken from each log were separated by about 5 m with the first section coming from near the base of the tree. Sections from old logs were kept intact by tightly wrapping them in plastic fencing material (square mesh, diameter: 4.5 cm) (Fig. 4B). All sections were transported by truck (~2.5 hr) to an emergence facility in

Athens, Georgia. Bark was lost from many of young log sections during transport. Emerging beetles were collected using rearing bags (Ulyshen and Hanula 2009) for about 8 months (8

August 2006 to 5 April 2007).

90 Beetles collected from both on- and off-location sampling were stored in 70% ethanol and later identified using the classification system of Arnett and Thomas (2001, 2002). Voucher specimens have been deposited in the Georgia Museum of Natural History, Athens, Georgia.

Statistical Analysis

ANOVA was used to compare species richness among decay classes for both on- and off- location sampling (SAS Institute 1990). Data from the different sampling periods for on-location sampling were combined before analysis. Similarly, data collected from the three sections removed from each log for off-location sampling were combined before analysis.

Species richness estimates were made using the Chao1 estimator, calculated as follows:

Chao1=Sobs + (a²/2b), where Sobs is the observed species richness, a is the number of singletons and b is the number of doubletons (Colwell and Coddington 1994). Estimates were made for the three stages of decay sampled off-location in this study as well as for four additional stages of decay sampled in a separate study. In that study (see Ulyshen and Hanula, in review), 0.5 m sections of loblolly pine were placed at various locations in plots containing different amounts of dead wood. The same emergence bags used in this study were used to collect beetles from the logs 2, 6, 10 and 22 months after the trees were killed. Beetles collected in the two studies were sorted and identified the same way by the same person (MDU). We estimated richness for each decay class instead of reporting observed richness values because sampling intensity differed between the two studies (i.e., 12 sections per decay class in this study compared to 36 sections per decay class in the other study).

Nonmetric Multidimensional Scaling was performed separately for the on- and off- location data using PC-ORD. After removing species found in fewer than three samples, the

91 datasets for on- and off-location sampling consisted of 85 and 40 species, respectively. PC-ORD was also used to perform indicator species analysis to identify species significantly associated with the different decay classes.

Results and Discussion

Methodological Considerations

Because on- and off-location sampling took place at different times of the year and involved wood at different stages of decay, it is not possible to meaningfully compare the efficiencies of the two methods. However, several benefits and drawbacks of the two methods became apparent during the course of this project. One advantage of off-location sampling was that it allowed us to control and accurately quantify the amount of material from which insects emerged, an important condition for experimentation (Carroll 1996). Although we measured the amount of wood enclosed within the traps used for on-location sampling, the exact volume of wood sampled is unknown as many beetles likely entered the trap by tunneling through the wood or under the bark. One disadvantage of off-location sampling was that it was relatively disruptive (i.e., removing the logs, transporting them by truck and hanging them vertically in the lab) compared to on-location sampling.

Data Set

Overall, we collected 10,506 beetles from 44 families and 209 species (Table 4.2).

Almost twice as many species were collected on-location than off-location (178 and 91 species, respectively). Because several abundant and species rich groups (e.g, Ciidae, Corylophidae,

Latridiidae, and Scydmaenidae) were not sorted below family level and other taxa were sorted

92 only to genus or to morphospecies, estimates of species richness presented in this paper are conservative. Females specimens of three pselaphine (Staphylinidae) genera were not identified to species and are treated as separate taxa from the identified male members of those genera

(Table 4.2)

Species Richness.

Observed and estimated species richness both peaked within the first year after death

(Figures 4.2, 4.3), most likely due to the diverse community of bark beetles and predators associated with that early stage of decay (Table 4.1). There were no differences in species richness among decay classes in the off-location sampling (Figure 4.2), probably because the young logs were then past the first stage of colonization and lacked most species associated with fresh wood (Table 4.2).

Although there have been few previous efforts to study the succession of insects in dead wood beyond the first two years of decomposition, there is some support for our conclusion that species richness declines steadily with time. For example, two observational studies on the succession of insects in Virginia pine in North Carolina and Maryland suggest that species richness declines during wood decay (Savely 1939; Howden and Vogt 1951). Furthermore, studies from Europe suggest that few species can live in wood toward the end of the decay process due to deteriorating nutritional quality (Siitonen 2001). Not all previous research is in agreement with our findings, however. Langor et al. (2008) contend that species richness increases with time in response to the diversification of wood-rotting fungi and increased microhabitat diversity. Support for this argument comes from Hammond et al. (2004), who found an increase in beetle species richness with time based on window trapping and rearing

93 from snags and logs of Populus in Canada. This discrepancy may be due in part to the fact that few phloeophagous species were collected in that study. Also, species richness may increase for a time following the phloem phase as logs become increasingly infiltrated by fungi and insects.

Some evidence for this comes from the mean species richness of beetles collected in off-location sampling (Figure 4.2) in this study. Considerably fewer species were collected in 15 month logs

(i.e., after the phloem phase) than in the two older decay classes.

Community Composition

The three decay classes formed distinct groupings for both on- and off-location sampling

(Figure 4.4). According to indicator species analysis, 25 and 7 species were significantly associated with young and old logs, respectively (Table 4.1). Only one taxon, Cossonus spp., was found to be significantly associated with middle-aged logs based on data from on-location sampling (Table 4.1). The significance of this association is in doubt, however, given that the same taxon was found to be significantly associated with young logs based on data from off- location sampling (Table 4.1). Most species associated with young logs are well known phloeophages (e.g., scolytines and cerambycids) and their predators. None of the species associated with old logs are known to be threatened and, to our knowledge, are not limited by poor dispersal abilities. However, there is an absence of information regarding the status and dispersal abilities of most species in the southeastern United States. Research on the relationship between dispersal ability and decay class association would be of great value. For instance, it would be interesting to compare the dispersal abilities of Corticeus thoracicus (Melsheimer) and

Dioedus punctatus LeConte, two similarly sized tenebrionids with different age-class associations (young and old logs, respectively, see Table 4.1).

94 Management Implications

It is clear from our results that species richness peaks within the first year of tree death due to a diverse community of phloem-feeders and their predators. These early colonists are thought to have excellent dispersal abilities and may be relatively resilient to forest management.

However, these species are also often highly host specific (Langor et al. 2008), underscoring the importance of preserving as much tree diversity as possible at the landscape scale.

Because species composition differed among decay classes and a number of species were significantly associated with the oldest logs sampled in this study, it is important to protect wood throughout the decomposition process in managed forests. The value of dead wood to conservation may be greatest at advanced stages of decay if, as discussed above, species associated with those stages are the most vulnerable due to poor dispersal powers.

Acknowledgements

We thank Chris Crowe, David White and Mike Cody for felling trees in 1997; Scott Horn and

Mike Cody for felling trees in 2005; Danny Dyer and Mike Cody for assembling emergence traps used for on-location sampling; Scott Horn, Mike Cody and Jared Swain for helping install traps and for assistance with collecting samples for on-location sampling; and Scott Horn and

Mike Cody for assistance with collecting and transporting wood samples for off-location sampling. We also thank Bob Rabaglia, Chris Carlton and Alexey Tishechkin for identifying scolytines, pselaphines and histerids, respectively, and Cecil Smith for assistance during visits to the Georgia Museum of Natural History. Support was provided by the Department of Energy-

Savannah River Operations Office through the U.S. Forest Service Savannah River under

Interagency Agreement DE-AI09-00SR22188.

95 Literature Cited

Arnett Jr., R.H., and Thomas, M.C. (Eds.), 2001. American Beetles: Archostemata, Myxophaga,

Adephaga, Polyphaga: Staphyliniformia. CRC Press, Boca Raton, Florida.

Arnett Jr., R.H., and Thomas, M.C. (Eds.), 2002. American Beetles: Polyphaga: Scarabaeiodea

through Curculionoidea. CRC Press, Boca Raton, Florida.

Barber, B.L., and Van Lear, D.H. 1984. Weight loss and nutrient dynamics in decomposing

woody loblolly pine logging slash. Soil Sci. Soc. Am. J. 48: 906-910.

Carroll, C.R. 1996. Coarse woody debris in forest ecosystems: an overview of biodiversity issues

and concepts In: McMinn, JW, and DA Crossley, Jr. (eds). Biodiversity and coarse

woody debris in southern forests, proceedings of the workshop on coarse woody debris in

southern forests: effects on biodiversity; 1993 October 18-20; Athens, GA. Gen. Tech.

Rep. SE-94. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern

Research Station. 146 p.

Colwell, R.K., Coddington, J.A. 1994. Estimating terrestrial biodiversity through extrapolation.

Phil. Trans. R. Soc. Lond. B 345: 101-118.

Dajoz, R. 2000. Insects and forests: The role and diversity of insects in the forest environment.

Intercept Ltd. New York. 668 p.

Elton, C.S. 1966. The Pattern of Animal Communities. Methuen and Co Ltd, London.

432 p.

Graham, S.A. 1925. The felled tree trunk as an ecological unit. Ecology 6: 397-411.

Hammond, H.E.J., Langor, D.W., and Spence, J.R. 2004. Saproxylic beetles (Coleoptera) using

Populus in boreal aspen stands of western Canada: spatiotemporal variation and

conservation of assemblages. Can. J. For. Res. 34: 1-19.

96 Harmon, M.E., Franklin, J.F., Swanson, F.J., Sollins, P., Gregory, S.V., Lattin, J.D., Anderson,

N.H., Cline, S.P., Aumen, N.G., Sedell, J.R., Lienkaemper, G.W., Cromack Jr., K., and

Cummins, K.W. 1986. Ecology of coarse woody debris in temperate ecosystems. Adv.

Ecol. Res. 15: 133-302.

Hines, J.W., and Heikkenen, H.J. 1977. Beetles attracted to severed Virginia pine (Pinus

virginiana Mill.). Environ. Entomol. 6: 123-127.

Howden, H.F., and Vogt, G.B. 1951. Insect communities of standing dead pine (Pinus virginiana

Mill.). Ann. Entomol. Soc. of Am. 44: 581-595.

Johansson, T., Gibb, H., Hilszczański, J., Pettersson, R.B., Hjältén, J., Atlegrim, O., Ball, J.P.,

and Danell, K. 2006. Conservation-oriented manipulations of coarse woody debris affect

its value as habitiat for spruce-infesting bark and ambrosia beetles (Coleoptera:

Scolytinae) in northern Sweden. Can. J. For. Res. 36: 174-185.

Kilgo, J.C., and Blake, J.I. (eds). 2005. Ecology and Management of a Forested Landscape: Fifty

years on the Savannah River Site. Island Press, Washington DC. 479 p.

Langor, D.W., Hammond, H.E.J., Spence, J.R., Jacobs, J., and Cobb, T.P. 2008. Saproxylic

insect assemblages in Canadian forests: diversity, ecology, and conservation. Can.

Entomol. 140: 453-474.

Moser, J.C., Thatcher, R.C., and Pickard, L.S. 1971. Relative abundance of southern pine beetle

associates in East Texas. Ann. Entomol. Soc. Am. 64: 72-77.

Nilssen, A.C. 1984. Long-range aerial dispersal of bark beetles and bark weevils (Coleoptera,

Scolytidae and Curculionidae) in northern Finland. Ann. Entomol. Fenn. 50: 37-42.

Nilsson, S.G., and Baranowski, R. 1997. Habitat predictability and the occurrence of wood

97 beetles in old-growth beech forests. Ecography 20: 491-498.

Overgaard, N.A. 1968. Insects associated with the southern pine beetle in Texas, Louisiana, and

Mississippi. J. Econ. Entomol. 61: 1197-1201.

SAS Institute, 1990. SAS Guide for Personal Computers. SAS Institute, Cary, NC, USA.

Savely Jr., H.E. 1939. Ecological relations of certain animals in dead pine and oak logs.

Ecol. Monogr. 9: 321-385.

Schiegg, K. 2000. Effects of dead wood volume and connectivity on saproxylic insect species

diversity. Ecoscience 7: 290-298.

Schultz, R.P. 1997. Loblolly Pine: the ecology and culture of loblolly pine (Pinus taeda L.).

Agricultural Handbook 713. USDA Forest Service, Washington, DC.

Siitonen, J., 2001. Forest management, coarse woody debris and saproxylic organisms:

Fennoscandian boreal forests as an example. Ecol. Bull. 49: 11-41.

Ulyshen, M.D., and Hanula, J.L., 2009. Habitat associations of saproxylic beetles in the

southeastern United States: A comparison of forest types, tree species and wood postures.

Forest Ecol. Manage. 257: 653-664.

Ulyshen, M.D., and Hanula, J.L., in review. Responses of arthropods to large-scale

manipulations of dead wood in loblolly pine stands of the southeastern United States.

Yee, M., Grove, S.J., Richardson, A.M.M., and Mohammed, C.L. 2006. Brown rot in inner

heartwood: why large logs support characteristic saproxylic beetle assemblages of

conservation concern. In: Grove, SJ and JL Hanula (eds.) Insect biodiversity and dead

wood: proceedings of a symposium for the 22nd International Congress of Entomology.

Gen. Tech. Rep. SRS-93. Asheville, NC: U.S. Department of Agriculture, Forest Service,

Southern Research Station. 109 p.

98 Table 4.1. Beetles species significantly associated with certain decay classes based on indicator species analysis.

Indicator Value P-value (on/off-location) Young log associates Acanthocinus nodosus (Fabricius) 25 (on) 0.025 Acanthocinus obsoletus (Olivier) 93.7 (on) 0.001 Cerylon unicolor (Ziegler) 31.2 (on) 0.006 Clavilispinus sp. 99.4 (off) 0.005 Colydium nigripenne LeConte 100 (off) 0.003 Colydium nigripenne LeConte 25 (on) 0.031 Corticeus thoracicus (Melsheimer) 99.9 (on) 0.001 Cossonus spp. 73.3 (off) 0.016 Dendroctonus terebrans (Olivier) 25 (on) 0.028 Diplocoelus rudis (LeConte) 36.4 (on) 0.021 Gnathotrichus materiarius (Fitch) 99.8 (on) 0.001 Ips calligraphus (Germar) 81.2 (on) 0.001 Ips grandicollis (Eichhoff) 37.5 (on) 0.003 Lasconotus pusillus LeConte 43.7 (on) 0.001 Monochamus titillator (Fabricius) 75 (on) 0.001 Myoplatypus flavicornis (Fabricius) 62.5 (on) 0.001 Myrmecocephalus sp. 46.3 (on) 0.001 Nacaeus tenuis (LeConte) 35.2 (on) 0.007 Nitidulidae sp. 9 43.7 (on) 0.002 Orthotomicus caelatus (Eichhoff) 55.8 (on) 0.001 Platysoma cylindricum (Paykull) 31.2 (on) 0.009 Platysoma parallelum (Say) 37.5 (on) 0.003 Plegaderus transversus (Say) 36.5 (on) 0.009 Thoracophorus costalis (Erichson) 82.1 (off) 0.031 Xyleborus ferrugineus (Fabricius) 52.7 (on) 0.001 Xyleborus pubescens Zimmermann 99.8 (on) 0.001 Middle-aged log associates Cossonus spp. 82.2 (on) 0.001 Old log associates Conoplectus canaliculatus (LeConte) 75 (off) 0.047 Dioedus punctatus LeConte 30.9 (on) 0.014 Eblisia carolina (Paykull) 93.3 (off) 0.012 Palaminus sp. 31.2 (on) 0.01 Philothermus glabriculus LeConte 80.4 (off) 0.043 Staphylinidae sp. 54 32.7 (on) 0.032 Uloma punctulata LeConte 88.6 (off) 0.016

99 Table 4.2. List of beetle species that emerged from loblolly pine off-location (i.e., in the lab) and on-location (i.e., in the field). The age of the logs, in months, is given at the top of each column.

Off-location On-location 15 60 108 1-10 46-55 94-103 Total Aderidae Cnopus impressus (LeConte) 0 2 16 1 0 1 20 Ganascus ventricosus (LeConte) 1 1 0 0 1 1 4 Pseudariotus notatus (LeConte) 0 1 0 0 0 0 1 Aderidae sp. 4 0 0 0 0 1 0 1 Anobiidae Anobiidae sp. 1 0 0 0 1 0 0 1 Biphyllidae Diplocoelus rudis (LeConte) 1 0 0 11 3 3 18 Buprestidae Buprestis lineata Fabricius 0 0 0 3 0 0 3 Chalcophora virginiensis Drury 0 5 1 0 1 2 9 Carabidae Calathus opaculus LeConte 0 0 0 1 0 0 1 Clivina pallida Say 0 1 0 0 0 0 1 Coptodera aerata Dejean 0 0 0 3 0 0 3 Cyclotrachelus laevipennis (LeConte) 0 0 0 0 1 1 2 Cymindis limbatus Dejean 0 0 0 0 1 0 1 Dicaelus ambiguus Laferte 0 0 0 1 0 1 2 Harpalus protractus Casey 0 0 0 0 0 1 1 Helluomorphoides clairvillei (Dejean) 0 0 0 1 0 0 1 Mioptachys flavicauda Say 54 37 31 0 0 3 125 Perigona nigriceps Dejean 0 0 0 1 0 0 1 Perigona pallipennis (LeConte) 0 0 0 1 0 0 1 Piesmus submarginatus (Say) 0 0 0 2 14 10 26 Polyderis laevis Say 0 0 0 1 1 0 2

100 Tachyta nana inornata (Say) 3 3 1 1 1 1 10 undetermined sp. 0 0 0 0 0 1 1 Cerambycidae Acanthocinus nodosus (Fabricius) 0 0 0 11 0 0 11 Acanthocinus obsoletus (Olivier) 0 0 0 89 0 0 89 Monochamus titillator (Fabricius) 0 0 0 41 0 0 41 Xylotrechus sagittatus (Germar) 1 0 0 1 0 0 2 Cerylonidae Cerylon unicolor (Ziegler) 0 0 0 5 0 0 5 Philothermus glabriculus LeConte 3 18 86 1 0 1 109 Chrysomelidae Chrysomelidae sp. 1 0 0 0 1 1 0 2 Chrysomelidae sp. 2 0 0 0 1 0 0 1 Ciidae Ciidae spp. 354 62 16 0 1 1 434 Cleridae Priocera castanea (Newman) 1 0 0 1 1 0 3 Thanasimus dubius (Fabricius) 0 0 0 1 0 0 1 Coccinellidae Harmonia axyridis (Pallas) 0 0 0 0 1 0 1 Colydiidae Bitoma carinata (LeConte) 0 0 0 1 0 0 1 Bitoma quadriguttata (Say) 0 0 0 0 3 0 3 Colydium nigripenne LeConte 37 0 0 17 0 0 54 Lasconotus pusillus LeConte 0 0 0 17 0 0 17 Namunaria guttula (LeConte) 1 0 0 2 0 0 3 Corylophidae Corylophidae spp. 1 1 0 0 2 0 4 Cryptophagidae Cryptophagidae sp. 2 0 0 0 3 3 0 6 Cryptophagidae sp. 3 0 0 0 0 1 0 1 Cryptophagidae sp. 4 0 2 0 10 3 3 18 Curculionidae Cercopeus sp. 0 0 0 0 0 1 1

101 Cossonus spp. 181 66 0 25 213 5 490 Dendroctonus terebrans (Olivier) 0 0 0 4 0 0 4 Dryophthorus americanus Bedel 10 40 1 0 5 1 57 Gnathotrichus materiarius (Fitch) 0 0 0 543 1 0 544 Hylastes salebrosus Eichhoff 0 0 0 3 0 0 3 Hylastes tenuis Eichhoff 0 0 0 0 0 1 1 Ips avulsus (Eichhoff) 0 0 0 1 0 1 2 Ips calligraphus (Germar) 0 0 0 89 0 0 89 Ips grandicollis (Eichhoff) 0 0 0 12 0 0 12 Myoplatypus flavicornis (Fabricius) 0 0 0 134 0 0 134 Orthotomicus caelatus (Eichhoff) 0 0 0 126 0 1 127 Pityophthorus spp. 0 0 0 0 0 1 1 Rhyncolus sp. 0 1 0 0 0 0 1 Xyleborinus saxesini (Ratzeburg) 0 0 0 1 0 2 3 Xyleborus ferrugineus (Fabricius) 10 0 0 30 1 1 42 Xyleborus pubescens Zimmermann 0 0 0 2525 0 4 2529 Xylosandrus crassiusculus (Motschulsky) 0 0 0 3 0 0 3 Derodontidae Derodontus esotericus Lawrence 0 0 0 0 3 0 3 Elateridae Alaus myops (Fabricius) 2 2 1 0 0 0 5 Ampedus areolatus (Say) 0 2 0 2 0 2 6 Ampedus luteolus LeConte 0 0 0 0 24 33 57 Athous cucullatus (Say) 0 0 0 3 7 3 13 Dicrepidius ramicornis (Palisot de Beauvois) 1 0 0 0 0 0 1 Dipropus spp. 1 2 4 1 2 12 22 Drapetes quadripustulatus Bonvouloir 1 0 0 0 2 0 3 Lacon impressicollis (Say) 0 2 0 1 0 0 3 Megapenthes rufilabris Germar 0 0 0 0 0 1 1 Melanotus ignobilis Melsheimer 0 1 3 98 1 1 104 Melanotus sp. 2 0 0 0 1 0 2 3

102 Neotrichophorus carolinensis Schaeffer 0 0 0 2 0 1 3 Orthostethus infuscatus Germar 0 0 0 0 0 1 1 Endomychidae Aphorista vittata (Fabricius) 0 0 0 7 6 3 16 Mycetina perpulchra (Newman) 0 0 0 0 2 0 2 Phymaphora pulchella Newman 0 0 0 0 1 0 1 Eucnemidae Dromaeolus sp. 0 2 0 1 1 0 4 Fornax sp. 0 0 0 0 0 11 11 Histeridae Aeletes floridae (Marseul) 0 1 0 0 0 0 1 Aeletes simplex (LeConte) 0 0 0 1 0 0 1 Bacanius punctiformis (LeConte) 57 13 17 0 0 2 89 Bacanius tantillus LeConte 0 4 24 0 0 0 28 Bacanius sp. 3 0 0 3 0 0 0 3 Caerosternus americanus (LeConte) 0 0 2 0 0 0 2 Eblisia carolina (Paykull) 0 1 14 1 0 2 18 Epierus pulicarius Erichson 2 8 7 0 0 0 17 Paromalus seminulum Erichson 0 1 17 0 0 2 20 Platysoma cylindricum (Paykull) 0 0 0 16 0 0 16 Platysoma parallelum (Say) 0 0 0 17 0 0 17 Plegaderus barbelini Marseul 0 0 0 1 0 0 1 Plegaderus transversus (Say) 0 0 0 38 1 0 39 Hydrophilidae Hydrophilidae sp. 1 0 0 0 1 0 1 2 Laemophloeidae Cryptolestes sp. 0 0 0 0 1 0 1 Laemophloeus biguttatus (Say) 0 0 0 1 0 0 1 Lampyridae Lampyridae sp. 1 0 0 0 1 0 0 1 Latridiidae Latridiidae spp. 0 0 0 4 10 4 18

103 Leiodidae Agathidium sp. 1 0 0 0 1 0 0 1 Agathidium sp. 2 0 1 5 2 0 2 10 Anisotoma sp. 0 0 0 0 0 2 2 Ptomaphagus sp. 0 0 0 1 0 0 1 Lycidae Dictyopterus aurora (Herbst) 0 0 0 0 0 1 1 Plateros spp. 0 25 12 0 24 15 76 Micromalthidae Micromalthus debilis LeConte 0 15 22 0 3 1 41 Monotomidae Bactridium sp. 0 0 0 2 0 0 2 Monotoma sp. 1 1 0 0 1 0 3 Rhizophagus cylindricus LeConte 0 0 0 2 0 0 2 Mycetophagidae Litargus tetraspilotus LeConte 0 0 0 1 1 0 2 Nitidulidae Conotelus sp. 0 0 0 1 0 0 1 Pallodes sp. 0 0 0 0 1 0 1 Thalycra sp. 0 0 0 0 0 1 1 Nitidulidae sp. 7 0 0 0 2 2 1 5 Nitidulidae sp. 8 0 0 0 2 0 0 2 Nitidulidae sp. 9 0 0 0 12 0 0 12 Passalidae Odontotaenius disjunctus (Illiger) 0 0 0 0 2 1 3 Passandridae Catogenus rufus (Fabricius) 1 2 0 2 0 0 5 Phalacridae Phalacridae sp. 3 0 0 0 0 1 2 3 Phalacridae sp. 4 0 0 0 1 2 0 3 Ptiliidae Ptiliidae spp. 11 6 0 0 0 0 17 Scarabaeidae Bolboceras thoracicornis (Wallis) 0 0 0 0 0 1 1 Euphoria sepulcralis (Fabricius) 0 0 0 1 0 0 1 Scraptiidae Canifa sp. 0 0 1 0 0 0 1 Scydmaenidae

104 Scydmaenidae spp. 1 21 20 1 7 8 58 Silvanidae Ahasversus rectus (LeConte) 0 0 0 3 1 1 5 Cathartosilvanus imbellis (LeConte) 0 0 0 1 0 1 2 Silvanus muticus Sharp 0 0 0 11 5 3 19 Uleiota dubius (Fabricius) 1 0 0 0 0 0 1 Sphindidae Sphindus sp. 0 3 8 0 0 5 16 Staphylinidae Acrolocha sp. 0 0 0 2 10 1 13 Actiastes sp. 0 0 14 0 0 0 14 Anacyptus testaceus (LeConte) 3 3 2 0 0 0 8 Batriasymmodes sp. 1 0 0 0 1 1 3 Batrisodes uncicornis (Casey) 0 0 0 0 0 3 3 Clavilispinus sp. 1731 10 0 1 55 4 1801 Conoplectus canaliculatus (LeConte) 0 0 27 0 0 0 27 Coproporus ventriculus (Say) 78 2 48 1 2 3 134 Dalmosanus sp. 0 0 1 0 0 0 1 Echiaster sp. 0 0 0 2 3 1 6 Euplectus duryi Casey 0 0 10 0 0 0 10 Euplectus sp. (female) 0 3 23 0 0 0 26 Gyrohypnus sp. 3 6 1 3 1 4 18 Hesperus sp. 0 0 0 0 0 1 1 Laetulonthus laetulus (Say) 0 0 0 1 0 0 1 Leptoplectus pertenuis (Casey) 0 0 0 0 1 3 4 Lordithon angularis (Sachse) 0 0 1 0 0 0 1 Lordithon obsoletus (Say) 0 0 0 0 0 1 1 Megalopinus caelatus (Gravenhorst) 0 0 0 0 0 1 1 Melba sp. (female) 0 1 0 0 0 0 1 Melba parvula (LeConte) 4 1 0 0 0 0 5 Melba sulcatula Casey 0 0 24 0 0 0 24 Myrmecocephalus sp. 0 0 0 14 2 1 17 Nacaeus tenuis (LeConte) 0 0 0 15 0 1 16 Oxyporus femoralis austrinus Horn 0 0 0 0 1 0 1 Palaminus sp. 0 0 2 2 1 15 20

105 Proteinus sp. 0 0 0 0 0 1 1 Pycnoplectus sp. (female) 0 1 0 0 0 0 1 Pycnoplectus interruptus (LeConte) 0 10 4 1 2 2 19 Pycnoplectus linearis (LeConte) 0 0 0 0 3 1 4 Pycnoplectus sexualis (Casey) 0 0 0 2 1 1 4 Scaphidiinae sp. 2 0 0 0 0 1 1 2 Scaphidiinae sp. 3 3 1 35 1 0 2 42 Scaphidium sp. 0 0 0 0 0 1 1 Sepedophilus scriptus (Horn) 0 0 152 0 5 10 167 Sepedophilus sp. 2 0 0 4 5 11 20 40 Sepedophilus sp. 3 0 0 2 0 5 8 15 Sepedophilus sp. 4 0 0 0 8 13 10 31 Stilicopsis sp. 0 0 0 0 1 0 1 Thesiastes pumilis (LeConte) 0 0 0 0 1 0 1 Thinocharis sp. 0 0 0 4 0 0 4 Thoracophorus costalis (Erichson) 23 5 0 0 0 2 30 Tmesiphorus costalis LeConte 0 7 14 0 1 0 22 Tyrus consimilis Casey 1 1 0 0 3 1 6 Staphylinidae sp. 29 0 0 7 11 1 27 46 Staphylinidae sp. 50 0 0 2 0 0 0 2 Staphylinidae sp. 51 0 0 0 1 0 0 1 Staphylinidae sp. 52 0 0 0 0 1 0 1 Staphylinidae sp. 53 0 1 9 0 0 0 10 Staphylinidae sp. 54 0 4 0 5 4 17 30 Staphylinidae sp. 55 0 1 0 0 0 0 1 Staphylinidae sp. 56 0 0 0 0 0 1 1 Staphylinidae sp. 57 1 0 0 11 1 3 16 Staphylinidae sp. 58 0 1 4 19 13 15 52 Staphylinidae sp. 59 0 0 1 0 0 0 1 Staphylinidae sp. 60 0 0 0 0 0 1 1 Staphylinidae sp. 61 1 0 0 0 0 0 1 Staphylinidae sp. 62 0 0 0 0 2 1 3 Staphylinidae sp. 63 0 0 0 1 0 0 1 Staphylinidae sp. 64 2 1 0 0 0 0 3 Staphylinidae sp. 65 0 0 1 0 0 2 3 Staphylinidae sp. 66 0 0 0 1 1 5 7 Tenebrionidae Corticeus parallelus (Melsheimer) 0 0 0 1 0 0 1

106 Corticeus thoracicus (Melsheimer) 0 0 0 914 0 1 915 Dioedus punctatus LeConte 0 13 69 0 1 91 174 Helops cisteloides Germar 0 0 0 1 2 0 3 Hymenorus sp. 0 0 30 4 3 5 42 Lobopoda erythrocnemis Germar 0 0 5 0 0 1 6 Platydema flavipes (Fabricius) 0 1 0 1 1 1 4 Platydema ruficorne (Stürm) 0 0 0 1 0 0 1 Poecilocrypticus formicophilus Gebien 0 1 0 0 0 0 1 Uloma punctulata LeConte 4 29 256 0 18 78 385 Tetratomidae Eustrophopsis bicolor (Fabricius) 0 0 0 1 0 0 1 Holostrophus bifasciatus (Say) 0 0 0 0 1 1 2 Throscidae Aulonothroscus convergens (Horn) 0 0 27 62 21 31 141 Aulonothroscus sp. 1 0 0 1 2 1 5 Zopheridae Hyporhagus punctulatus Thomson 0 0 0 6 0 0 6 Pycnomerus haematodes (Fabricius) 20 37 0 4 3 1 65 Pycnomerus sulcicollis LeConte 19 23 21 18 14 15 110 Total individuals (species) 2632 516 1108 5094 584 572 10506 (41) (58) (52) (112) (86) (102) (209)

107 Figure 4.1. Emergence bags used for off-location sampling (A). Old logs sampled off-location were held together with plastic fencing material (B). Emergence trap used for on-location sampling (C).

108 109 Figure 4.2. Mean ± SE number of beetle species collected from different aged loblolly pine logs using emergence traps on-location (i.e., in the field) and off-location (i.e., in the lab). Within each graph, means with different letters next to them are significantly different based on Tukey’s

Studentized Range Test. Note: This figure should not be used to compare the efficiencies of the two trapping methods due to differences in sampling intensity (see methods).

110 35 On-location Sampling (n=16) Off-location Sampling (n=4)

30

A 25 A A

20 A

15 B B

Mean Numberof Species 10 1-10 46-55 94-103 15 60 108 Time after Death (Months)

111 Figure 4.3 Estimated number of beetle species in loblolly pine logs at different stages of decay on the Savannah River Site, South Carolina.

112 120

100

80

60 Estimated Number of Species

0 20 40 60 80 100 Time after Death (Months)

113 Figure 4.4. Nonmetric multidimensional scaling plots for on-location sampling (a 3-dimensional solution) and off-location (a 2-dimensional solution). Symbols represent young logs (solid circles), middle-aged logs (open circles) and old logs (triangles).

114 On-Location Sampling Off-Location Sampling

2 1

1 Axis 2

0 0 Axis 3 Axis

2 -1

1

-2 -1 0 Axis 1 -2 -1 0 1 2 1 0 Axis 1 -1 Axis 2 -1 -2

115 CHAPTER 5

CONCLUSIONS

Dead wood continues to be viewed by many private landowners as well as many professional foresters as forest waste instead of as critical wildlife habitat. This perception must be changed if saproxylic species are to be protected long-term in increasingly modified landscapes. One impediment to convincing landowners to protect dead wood is the widely-held belief that it benefits harmful pest species. Although major inputs of dead wood have been shown to briefly increase the outbreak risk of the European bark beetle, Ips typographus, in parts of Europe (Schroeder and Lindelöw 2002), this does not appear to be the case for any pest species in the southeastern United States. Because it supports a diverse community of parasitoids, predators and competitors of many pest species, dead wood may actually decrease the likelihood of pest outbreaks in many forests (Martikainen et al. 1999, Bouget and Duelli

2004, Coyle et al. 2005, Johansson et al. 2007). Losey and Vaughan (2006) recently estimated that control of crop pests by insects is worth $4.5 billion per year in the United States. Although such estimates have not been made specifically for pest control in forested settings, the economic benefits that come from having a diverse community of saproxylic insects may be substantial.

People are beginning to recognize the economic value of intact ecosystems (Holling and Meffe

1996) and restoring dead wood is an essential step in efforts to protect biodiversity in managed forests.

116 Literature Cited

Bouget, C., Duelli, P. 2004. The effects of windthrow on forest insect communities: a

literature review. Biological Conservation 118: 281-299.

Coyle, D. R., Nebeker, T. E., Hart, E. R., Mattson, W. J. 2005. Biology and management of

insect pests in North American intensively managed hardwood forest systems. Annual

Review of Entomology 50: 1-29.

Holling, C. S., Meffe, G. K. 1996. Command and control and the pathology of natural resource

management. Conservation Biology 10: 328-337.

Johansson, T., Gibb, H., Hjältén, J., Pettersson, R. B., Hilszczański, J., Alinvi, O., Ball, J. P.,

Danell, K. 2007. The effects of substrate manipulations and forest management on

predators of saproxylic beetles. Forest Ecology and Management 242: 518-529.

Losey, J. E.,Vaughan, M. 2006. The economic value of ecological services provided by insects.

Bioscience 56: 311-323.

Martikainen, P., Siitonen, J., Kaila, L., Punttila, P., Rauh, J. 1999. Bark beetles (Coleoptera,

Scolytidae) and associated beetle species in mature managed and old-growth boreal

forests in southern Finland. Forest Ecology and Management 116: 233-245.

Schroeder, L. M., Lindelöw, Å. 2002. Attacks on living spruce trees by the bark beetle Ips

typographus (Col. Scolytidae) following a storm-felling: a comparison between stands

with and without removal of wind-felled trees. Agricultural and Forest Entomology 4:

47-56.

117