This item has the following downloads:
1 ENVIRO NM E N TAL CHANGES AFFECTING DOMINANT ANT SPECIES By JULIAN RESASCO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Julian Resasco
3 To my family mentors, and friends
4 ACKNOWLEDGMENTS I am foremost thankful to have had Doug Levey as my PhD advisor. Doug trained me as a scientist and truly taught me what it means to be a great mentor. Rob Fletcher has been like a second advisor to me and is an inspiration as a scientist and teacher. I am grateful to everyone who has served on my committee: Sanford Porter for his ge nerosity in sharing his time experience, and lab space, Todd Palmer for many good conversations about research, Ben Bolker for his devotion to training students in quantitative skills, Scott Robinson, for research feedback at so many Lunc hbunch es I thank the USDA Forest ServiceSavannah River for their collaboration and support. I am especially thankful to John Blake, who went above an d beyond to facilitate my research. I feel very fortunate to have had the pleasure to work with the great people invo lved in the Corridor Project at Savannah River Site (SRS) I thank Nick Haddad, Lars Brudvig, John Orrock, Ellen Damschen, and Josh Tewksbury for being mentors and role models. Candice Hardwick, Camille Beasley, and Viviana Penuela provided assistance in t he field. For logistical help at SRS I thank Ed Olson, Jamie Scott and Kim and Pat Wright Lars Brudvig, Lauren Sullivan, Dan Evans, Christine Brown, Chris Habeck, Nash Turley, Cathy Collins and, Joe Ledvina made SRS a good environment for academic devel opment. All Corridor Project technicians an d graduate students made South Carolina a great place to work, explore, and have fun. I thank the USDA ARS, Gainesville, FL for collaboration. DeWayne Shoemaker, generously agreed to do genetic analyses on numerous fire ant samples. Sanford Porter generously provided lab space and materials for experiments Eileen Carroll, and Darrell Hall provided lab assistance and expertise.
5 I am also grateful for mentorship fr om scientists peers, and teachers that I have had the pleasure to interact with and learn from : Nate Sanders, Rob Dunn, Diego Vzquez, Nachu Chacoff, DeWayne Shoemaker, Shannon Pelini, Emilio Bruna, Lloyd Davis, Katie Stuble, Andrew Hein, Adrian Stier, Al ex Jahn, Jill Jankowski Connie Clark Jackson Freshette, Gustavo Londoo, Ari Martinez Ashley Seifert, Megan Gittinger Seifert, Kathleen Rudolph, Andrew Hein, Miguel Acevedo, Divya Vasudev Brian Reichert, Trevor Caughlin, Dave Armitage, and Daniel Sasso n. Funding was provided by National Science Foundation (NSF Graduate Research Fellowship under Grant No. DGE 0802270, DEB 0614333, and REU supplement), funds provided to the Department of Agriculture, Forest ServiceSavannah River (a National Environmental Research Park), under Interagency Agreement DE AI09 00SR22188 with the Department of Energy, Aiken, SC, USDA ARS, Gainesville, FL, Southeast Alliance for Graduate Education and the Professoriate (University of Florida [UF]), IGERT QSE3 research grant, Department of Biology (UF), and College of Liberal Arts and Sciences (UF).
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 ABSTRACT ..................................................................................................................... 9 CHAPTER 1 INTRODUCTION .................................................................................................... 11 2 HABI TAT CORRIDORS ALTER RELATIVE TROPHIC POSITION OF FIRE ANTS ...................................................................................................................... 18 Methods .................................................................................................................. 21 Results .................................................................................................................... 24 Discussion .............................................................................................................. 25 3 LANDSCAPE CORRIDORS CAN INCREASE INVASION BY AN EXOTIC SPECIES AND REDUCE DIVERSITY OF NATIVE SPECIES ............................... 32 Methods .................................................................................................................. 34 Results .................................................................................................................... 39 Discussion .............................................................................................................. 41 4 TESTING SODIUM L IMITATION OF FIRE AN TS IN THE FIELD AND LABORATORY ....................................................................................................... 50 Methods .................................................................................................................. 51 Results .................................................................................................................... 54 Discussion .............................................................................................................. 55 5 USING HISTORICAL AND EXPERIMENTAL DATA TO REVEAL WARMING EFFECTS ON ANT COMMU NITIES ....................................................................... 60 Methods .................................................................................................................. 61 Results .................................................................................................................... 63 Discussion .............................................................................................................. 65 6 SUMMARY AND CONCLUSIONS .......................................................................... 72 LI ST OF REFERENCES ............................................................................................... 76 BIOGRAPHICAL SKETCH ............................................................................................ 87
7 LIST OF TABLES Table page 1 1 Geographic coordinates of corridor experimental landscapes (blocks). ............. 16 3 1 Species list and abundances. ............................................................................. 43 3 2 Diversity measures of native ants by patch types ............................................... 45 5 1 Species list for Savannah River Site and Duke Forest. ...................................... 67
8 LIST OF FIGURES Figure page 1 1 Photograph of one of ten experimental landscapes. ........................................... 17 2 1 Corridor experimental landscapes at Savannah River Site, SC. ........................ 28 2 2 Corridor and patch shape effects on fire ant mean and range of 15N. ............... 29 2 3 Relationship between species richness of plants and mean 15N of fire ants.. ... 30 2 4 Relationship between patch fire ant abundance and mean fire ants 15N. ......... 31 3 1 Aerial photograph and layout of one block (n = 8 blocks). .................................. 46 3 2 Effect of patch type on fir e ant abundance, measured as proportion of traps with fire ants and number of nests per patch for patch types. ............................ 47 3 3 Effect of c orridors on fire ant abundance and species richness and the relationship between fire ant abundance and species richness.. ........................ 48 3 4 Species richness and evenness (Hurlberts PIE) sample based rarefaction curves for patch types.. ...................................................................................... 49 4 1 Photograph of one exper imental fire ant colony ................................................. 56 4 2 Fire ant response to NaCl baits in Florida and South Carolina. .......................... 57 4 3 NaCl treatment effects on experimental fire ant colonies. .................................. 58 4 4 Sodium content of combined samples of fire ants from each of the four NaCl supplementation treatments. .............................................................................. 59 5 1 Annual, summer, and winter mean monthly temperatures near Savannah River Site, SC 1975 2011. .................................................................................. 69 5 2 Relatio nships between temperature and species richness, evenness, and turnover at Savannah River Site and Duke Forest. ............................................ 70 5 3 Relationship between temperature and species relative abundances for ants species that occurred at both Savannah River Site and Duke Forest. ................ 71
9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ENVIRONMENTAL CHANGES AFFECTING DOMINANT ANT SPECIES By Julian Resasco August 2013 Chair: Doug Levey Major: Zoology Habitat loss and fragmentation, species invasion, and climate change are the gravest threats to biodiversity. A major challenge is to predict how these threats will affect species. Ants are ecologically important, abundant, and indicators of ecosystem health and biodiversity of other taxa. They are also amenable to experimentation and so are good study organisms for understanding the effects of fragmentation, invasion, and climate change. In this dissertation, I used experiments to address questions related to these threats and their effect on ants: (1) How do fragmentation and corridors affect the trophic position of fire ants ( Solenopsis invicta)? (2) How do fragmentation and corridors af fect the invasion of fire ants and communities of native ants? (3) How does sodium availability affect behavioral response to salt and colony growth in fire ants? (4) How do warming temperatures affect ant communities of southeastern North American oak for ests? I found that: (1) fire ants in habitat patches connected by corridors had a higher estimated trophic position than fire ants in isolated patches. Because fire ants are generalist s, this suggest s that habitat fragmentation has negative effects on trophic
10 structure and that corridors help to ameliorate those negative effects. (2) Polygyne fire ants in connected patches had a higher abundance than polygyne fire ants in unconnected patches, while monogyne fire ants did not differ between patch types. As a result, in landscapes with polygyne fire ants, native ant diversity in connected patches was lower than in unconnected patches. This suggests that corridors can have the negative effect of spreading invasive species to the detriment of native species, but whether they do depends on species traits (3) Fire ants in the field in regions of low sodium deposition had a tenfold greater recruitment r esponse than fire ants from regions of high sodium deposition. In the laboratory, reared colonies did not show signs of sodium limitation on colony growth. (4) Ants under historical warming and experimental warming differed at the community level, however, some species showed similar responses to warming (positive and negative) in both sites These results suggest that ant species responses to warming temperatures are variable.
11 CHAPTER 1 INTRODUCTION Although humans have been affecting the environment for millennia, their impact has been especially great in the last century due to population growth and technology (Vitousek et al. 1997). As human pressure for resources has intensified, the proportion of Earths land area allocated to meeting those demands has grown. Almost half the area of terrestrial ecosystems has been transformed for human use (Vitousek et al. 1997); agricultural and pastoral lands are now the most extensive ecosystem in the world (Foley et al. 2007). In conjunction with increased commerce and travel, this land modification has led to parallel increase invasive species These invasive species are considered the second gravest threat to conservation, trailing only habitat destruction and degradation in importance (Wilcove et al. 1998). Anot her major driver of environmental change is climate change. Emissions from burning fossil fuels are changing the composition of the atmosphere and are changing global climate (IPCC 2007). Climate change has been documented to cause shifts in range and phen ology and extinctions of plants and animals (Parmesan 2006). A leading challenge for ecologists and evolutionary biologist is to predict how these environmental changes will affect biota. In spite of the prevalence of habitat alteration, species invasion, and climate change, their effects are often unclear because they are c onfounded with other factors, making mechanisms challenging to deduce. Controlled, replicated experiments can provide opportunities to better understand the effects of these threats. A nt communities provide an ideal study system to assess impacts of environmental change. Ants are diverse, ecologically important, and found throughout
12 the world (Hlldobler and Wilson 1990). They are indicators of the diversity of other taxa (Alonso 2000) and environmental health (Whitford et al. 1999; Andersen et al. 2002; Ellison 2012), are amenable to experimentation and include some of the most widespread and problematic invasive species (Moller 1996). Finally, ant communities often have a highly a skew ed distribution of species abundance, such that one or a few species are numerically dominant. Because these dominant species typically have a strong influence on other species in the community, they provide a natural focus for studies of how environmental change may drive ecological shifts in ant communities In this dissertation, I investigate how human changes to the environment influence dominant ants and ant communities. Three chapters focus on fire ants, Solenopsis invicta a dominant invasive species that suppresses and displaces other species (Porter and Savignano 1990; Gotelli and Arnett 2000; Tschinkel 2006), and one chapter focuses on the effects of warming temperatures on ant communities, and in particular the species Crematogaster lineolata, a dominant forest species. Data from this dissertation will be archived in Knowledge Network for Biocomplexity after publication. In C hapter 2 I investigated how fragmentation and corridors strips of habitat that connect otherwise isolated habitat patches affect the trophic position of fire ants. In largescale, experimental landscapes at Savannah River Site, SC (Fig. 1 1 ; Table 11 ) I collected fire ants from patches connected by corridors and from isolated patches. Nitr ogen stable isotopes, which can be used to reveal trophic position, showed that fire ants in connected patches had a higher average and range of values, signifying higher average trophic position and trophic breadth, respectively. Because fire ants are
13 gen eralist c onsumers, these results suggest that corridors help alleviate effects of habitat fragmentation on trophic structure. In Chapter 3 I used the same largescale experiment as in Chapter 2 to determine whether corridors affect fire ant invasion by f ire ants and species diversity of native ants. I did so by systematically collecting both types of ants in connected and unconnected patches. I found that dispersal traits determined by fire ant social form determined the effect of corridors on fire ant abundance and ant diversity. The two social forms are polygyne, where queens disperse poorly but establish at high densities, and monogyne, where queens disperse well but establish at low densities. Typically, all fire ants in the experiment al replicates wer e either all polygyne or all monogyne. In replicates with the polygyne social form fire ants tended to be more abundant and native ant diversity lower in connected than unconnected patches. In contrast, replicates with monogyne social form showed no difference in fire ant abundance or native ant diversity. These results suggest that corridors can have the unintended consequence of spreading at least some types of invasive species and that variation in dispersal ability within and among species can be used to predict the threat of that spread. In Chapter 4 I tested the effect of environmental sodium deposition on the behavioral response of fire ants to salt baits and determined the effect of supplemental sodium on fire ant colony growth. Sodium is an essent ial dietary element (Frausto da Silva & Williams 2001), and preferential foraging for high concentrations of sodium by inland herbivorous and omnivorous ants suggests it may be limiting (Kaspari 2008; 2009). If so, increased sodium availability through al tered deposition and anthropogenic sources may lead to increased colony growth and cascading ecological impacts. To test
14 the effect of so dium deposition on the behavior of fire ants, I collected ants at salt baits of different concentrations along transect s in regions of naturally high and low sodium deposition. Fire ants in low sodium deposition regions responded about an order of magnitude more strongly to high concentrations of NaCl baits than fire ants in high sodium deposition. To test the effect of su pplemental sodium on fire ant colony growth, I reared fire ant colonies in the laboratory with identical diets but varying concentrations of salt. Fir e ants grown in the laboratory did not show signs of sodium limitation on colony growth In Chapter 5, I investigated possible effects of warming temperature on the ant community structure of southeastern North American oak forests. Historical records of biotic communities can be compared to current records to suggest effects of recent climate change (Sagarin et al 1999; Smith et al 2006; Damschen et al. 2010; reviewed in Parmesan 2006). However, attribution of biotic changes to climate changes can be problematic due to confounding variables. Experiments that manipulate projected climates can overcome this i ssue of attribution, but long term implications of observed community responses from these typically short term experiments are uncertain. Combining both approaches can provide a powerful approach to reveal the effects of climate change on biological communities. I combined observational and experimental data to explore potential effects of warming temperatures on ant communities. Observational data came from a comparison of historical ant community data ( 1976) to contemporary data (2010 and 2011) within th e same oak forest stands, over which time temperatures increased approximately 2.7 C. Experimental data came from a warming manipulation at Duke Forest in which temperatures were increased 1.5 5.5 C above
15 ambient (Pelini 2011 a ). I found decreases in s p ecies richness and evenness in the observational study but not the experimental study. Under both experimental and natural warming, some species responded in a similar way; most notably, the abundance of the dominant ant, C. lineolata, increased.
16 Table 11 Geographic coordinates of corridor experimental landscapes (blocks). Block names correspond to USDA Forest ServiceSavannah River timber compartments. Block name Latitude Longitude 8 33.36988 81.70264 10 33.29907 81.74933 52 33.31133 81.59084 53N 33.32538 81.58588 53S 33.31085 81.57322 54N 33.32302 81.53546 54S 33.29371 81.52915 57 33.27925 81.52286 75E 33.19876 81.55293 75W 33.19196 81.56986
17 Figure 11. Photograph of one of ten experimental landscapes (photo courtesy of Ellen Damschen)
18 CHAPTER 2 HABITAT CORRIDORS ALTER RELATIVE TROPHIC POSITION OF FIRE ANTS The alteration of area, shape, and isolation of habitat patches can drive changes in species composition, biotic processes, and trophic struct ure (Kruess and Tscharntke 1994, Gilbert et al. 1998, Fahrig 2003, Collinge 2009) Such changes are increasingly common as natural landscapes become subdivided by anthropogenic processes. Corridors are a commonly implemented management strategy to mitigat e the negative effects of fragmentation because they facilitate movement of organisms between otherwise isolated patches (Hilty et al. 2006) Indeed, many studies have demonstrated that corridors generally increase rates of inter patch movement (Sutcliffe and Thomas 1996, Gonzalez et al. 1998, Tewksbury et al. 2002, Haddad and Tewksbury 2005, Gilbert Norton et al. 2010, but see Hilty et al. 2006 and Haddad et al. 2011a) These increases in movement are linked to increased species richness (Gilbert et al. 1998, Damschen et al. 2006, Damschen et al. 2008) and greater persistence of predators in connected habitat patches (Gilbert et al. 1998). However, the impact of corridors on community and trophic structure is still not well understood. Determining how chang es in landscape characteristics affect trophic structure is important, given the central role of trophic dynamics in ecosystem stability and function (McCann 2000, Estes et al. 2011) Fragmentation may alter trophic st ructure via several mechanisms. Theory predicts a positive relationship between the number of trophic levels and habitat area (Schoener 1989, Holt 1993, Pimm 2002) a pattern supported by nonexperimental, empirical studies (Vander Zanden et al. 1999, Komonen et al. 2000, Post et al. 2000, Lay man et al. 2007a, Takimoto et al. 2008, McHugh et al. 2010) Although such studies provide insight on potential effects of habitat loss, confounding
19 factors make it difficult to pinpoint the underlying mechanisms. Studies that experimentally manipulate fragmentation and connectivity to test how they affect organisms of different trophic levels are rare (Gilbert et al. 1998, Holyoak 2000, Davies et al. 2001) A particular challenge in fragmentation and corridor studies is to separate connectiv ity effects f rom edge effects. The challenge arises because corridors essentially always increase the edgeto area ratio of associated patches, and edges are well known to alter species abundance, distribution, interspecific interactions, and ecosystem processes (Harri son and Bruna 1999, Davies et al. 2001, Laurance et al. 2002, Ries et al. 2004) This confounding of connectivity and edge effects raises the question to what extent could differences in trophic structure between habitat patches with and without corridor s be due to connectivity versus differences in patch shape (i.e., edgeto area ratio)? An equally difficult challenge is quantification of the t rophic structure of a food web. A potential solution is provided by stable isotopes; recent work suggests that s table isotope ratios of generalist consumers can reflect fragmentation effects on trophic structure (see below, Layman et al. 2007a). In a largescale experiment, we quantified corridor and patch shape effects on trophic structure. We did so by quantifying indirect measures of trophic position and dietary breadth for a generalist consumer, the fire ant ( Solenopsis invicta Buren). We used stable isotopes -the ratio of 15N/14N (hereafter 15N), to estimate the relative trophic posi tion of fire ants between patch types (Bearhop et al. 2004, Newsome et al. 2007) In particular, we use d mean 15N of colonies in a patch to estimate fire ant trophic position (higher 15N indicates higher trophic position) and range (maximum
20 minus minimu m values of 15N within a patch) to estimate of range of trophic positions (hereafter trophic breadth). Fire ants are ideal study organisms for answering our questions about whether corridors alter trophic structure because they are trophic generalists (T schinkel 2006) As such, the 15N of a fire ant colony likely reflects the average trophic position of nearby prey items and 15N of a population of colonies in a habitat patch can be interpreted to reflect the trophic structure of that patch (Layman et al 2007a) We address three questions. (1) Do corridors increase trophic position and trophic breadth of fire ant colonies? Because corridors increase the movement of organisms (Gilbert Norton et al. 2010) help sustain species richness (Gilbert et al. 1998, Damschen et al. 2006) and allow greater persistence of predators in connected habitat patches (Gilbert et al. 1998) we predicted that fire ant colonies in unconnected patches would have a lower mean trophic position and narrower trophic breadth tha n those from patches connected by a corridor but otherwise similar in shape (i.e., edgeto area ratio). Because our experimental design also allows us to test for patch shape effects while controlling for connectivity, we also asked: (2) Does patch shape affect trophic position and trophic breadth of fire ants? To address possible mechanisms underlying connectivity effects, we ask: (3) Does plant species richness positively correlate with the mean 15N of fire ants across patches? Plant species richness is enhanced by connectivity (Damschen et al. 2006) and can increase abundance of predatory arthropods (Haddad et al. 2009) If corridors indeed increase the trophic position of fire ants (question 1), a positive relationship between plant species richness and 15N of fire ants would suggest that plant species richness is likely affecting the
21 trophic structure of consumers (i.e., the prey of fire ants) and ultimately influencing the higher trophic position of fire ants in connected patches. Methods St udy s ite We conducted this study at the Savannah River Site, South Carolina (33.20N, 81.40W) in ten randomized and replicated blocks designed to test effects of corridors on ecological processes. Patches were created by clearing mature pine plantation forest and are now managed for restoration to longleaf pine savanna, a species rich, endangered habitat (Van Lear et al. 2005, Jose et al. 2006) Restoration practices include prescribed fire, planting of longleaf pine ( Pinus palustris ) seedlings, and mechani cal rem oval of hardwood trees. As a result, the ground layer vegetation in patches is diverse and productive relative to the understory of the pine forest matrix. Each block consists of five patches. In the center of each block is a 1 ha center patch, and four peripheral patches. Peripheral patches are 150 m from the center patch and are of three types: connected, rectang ular, and winged. Connected patches are connected to the center patch by a 150 m long, 25m wide corridor of the same habitat (Fig. 2 1 A) Rectangular and winged patches lack a corridor connecting to the center patch and are thus isolated or unconnected. They are equal in area to the connected patch, 1.375 ha (1 ha plus the area of the corridor, 0.375 ha). The equivalent area of the corridor is added to the winged patches as two lateral 75 m long, 25 m wide wings and to the rectangular patches as additional area on the far side of the patch. Connected and winged patches have similar edgeto area ratios (491 m/ha and 509 m/ha, respectivel y) which are substantially higher than the edgeto area ratio of rectangular patches (345 m/ha). This design allows us to test for corridor effects independent of patch shape effects by comparing response variables from connected
22 patches to those from wing ed patches, which are unconnected but nearly identical in edgeto area ratio. Similarly, we can test for patch shape effects independent of corridor effects by comparing response variables from winged patches to those from rectangular patches, which differ greatly in edgeto area ratio but are identical with respect to connectivity (i.e., they are both unconnected). Each block contains either a second winged or a second rectangular patch (Fig. 2 1B). We averaged response variables from duplicate patch types within each block. Study organism, sampling, and stable isotope a nalysis Fire ants are opportunistic, omnivorous feeders that consume a wide variety of invertebrate prey, scavenged vertebrates, small seeds, plant exudates, and homopteran honeydew. Their diet typically reflects immediately available food sources rather than strong dietary preferences (Tschinkel 2006) Depending on resources available, they can be highly carnivorous (Tillberg et al. 2007) or omnivorous (Lofgren et al. 1975) Stable isotopes are frequently used to characterize food webs and estimate trophic position of consumers (Layman et al. 2007b, Newsome et al. 2007, Schmidt et al. 2007) 15N is particularly useful because nitrogen from the tissue of consumers is enriched in 15N relative to that of prey, which means that 15N is positively correlated with a consumers trophic position (Gannes et al. 1998, Post 2002) 15N values of consumers are often calibrated against 15N baseline values of plants. We did not make baseline adjustments. However, a recent study suggests that at spatial scales comparable to our withinblock comparisons, unadjusted 15N of consumers result in reliable comparisons of relative trophic position (Woodcock et al. 2012).
23 The enrichment in 15N for one consumer trophic level transfer (including insects) is generally 34 (Mooney and Tillberg 2005, Tillberg et al. 2006) In their native range, fire ant nests are highly variable in 15N ratios. This variance reflects an estimated span of two trophic level s (Tillberg et al. 2006) Here we use the patch mean 15N, which estimates the average trophic position of fire ant colonies in a given patch and patch range (maximum minus minimum) of 15N, which estimates the breath of trophic positions of fire ant colo nies in a given patch. Both mean and range of 15N have been used previously to characterize trophic structure in fragmented landscapes (Layman et al. 2007a) and mean 15N has been used to infer trophic position and prey availability in ants (Blthgen et al. 2003, Palmer 2003, Tillberg et al. 2007, Gibb and Cunningham 2011) In each peripheral patch (connected, rectangular, and winged) we collected fire ants using a stratified design. We divided each patch into a 3x3 sampling grid (N = 9 cells, each 33 .3m x 33.3m; Fig. 2 1B). We attempted to collect fire ants from one nest per cell w ithin the grid in all patches. When we did not find nests within a cell, we collected whenever possible from the nearest fire ant nest in an adjacent cell with the same amount of edge habitat. From each nest, we collected approximately 200 wor kers and stored them at 17C. Of these, we haphazardly selected 3050 individual s for stable isotope analysis. To obtain an estimate from a relatively long time window of nit rogen assimilation and unbiased by the last meal ingested, we analyzed only heads and thoraxes (Tillberg et al. 2006) Samples from a given nest were thoroughly mixed, weighed to the nearest + 1g, and analyzed for 15N at the University of Georgia Savannah River Ecology Laboratory using continuous flow isotope ratio a mass
24 spectrometer (Finnigan Delta plusXL; FinniganMAT, San Jose, CA). We report stable isotope ratios in per mil units () in the standard delta ( ) notation. We tested for normality of res ponse variables in each patch type using a ShapiroWilk test. Response variables did not require transformation. Data on plant species richness were collected via visual censuses over the entire area of all patches in eight blocks (Damschen et al. 2006) To determine whether trophic position was increased by corridors (question 1) we tested a directional hypothesis using paired, onetailed t tests on the mean and range of 15N from winged vs. connected patches. To determine whether trophic position was aff ected by patch type (question 2) we tested a nondirectional hypothesis using paired, twotailed t tests on the mean and range of 15N from winged vs. rectangular patches. One block was omitted from the range of 15N analysis because one patch type within that block was insufficiently sampled. To examine the relationship between plant species richness and trophic position of fire ants (question 3), we used a linear regression of plant species richness and mean 15N of fire ants from the same patch. All analyses were conducted using R (R Development Core Team 2009) Results We collected fire ants from 322 nests, with a mean of 8.05 ( 2.17 SD) nests per patch. 15N from these nests varied substantially ranging from 1.03 to 4.13. Mean 15N values were ~10% higher in connected than in unconnected patches of similar shape ( 15N = 1.86 and 1.69 in connected and winged patches, respectively; t = 2.10; P = 0.03), supporting our prediction (Fig. 2 2). Likewise, the range of 15N was ~33% greater in connected than unconnected patches of similar shape (ranges = 1.70 and
25 1.28 in connected and winged patches, respectively; t = 1.95; P = 0.04; Fig. 2 2 ), as predicted. Patch shape did not affect the trophic s ignature of fire ant col onies. In particular, mean and range of 15N did not differ between winged and rectangular patches, which differ greatly in edgeto area ratio but are identical in size and connectivity (t = 1.13; P = 0.29 for mean; t = 0.72; P = 0.50 for range; Fig. 22 ). The relationship between plant species richness and 15N of fire ant colonies was highly dependent on patch type (Fig. 2 3). In connected and rectangular patches the relationship was strong and positive (r2 = 0.60 and 0.52, respectively; P < 0.05 for both regressions ), whereas in winged patches there was no relationship between plant richness and mean fire ant 15N (r2 = 0, P > 0.94). We also explored rela tionship between fire ant 15N and fire ant abundance (see Chapter 3) and found no apparent pattern (Fig. 24). Discussion Our results suggest that patch connectivity (but not patch shape) has an effect on relative trophic position and br eadth of a generalist consumer. As predicted, fi re ant nests in connected patches had lower mean and range of 15N than those in similarl y shaped, unconnected patches. While the mechanism driving this pattern is uncertain, the positive correlation between patch plant species richness and mean fire ant 15N in most patch types suggests that plant diversity may play a role, since plant species richness is higher in connected than unconnected patches (Damschen et al. 2006). Due to the wide range of resources consumed by fire ants, the dietary shifts we observed may reflect local trophic structure of the habitat patches in our study sys tem
26 (see Layman et al. 2007a). Assuming that fire ant diets reflect local food sources, the lower 15N in isolated (winged) patches could signify an overall decrease in av ailability of animal prey, a decrease in trophic position of those prey or both. Likewise, the lower 15N range in isolated patches suggests that fire ants in those patches occupy a narrower range of trophic positions th an those in connected patches. While we are unable to discern whether corridors increased the abundance or trophic position of prey, our results nevertheless suggest that isolation negatively impacts prey and that corri dors mitigate such disruptions. Further work of this kind using focal org anisms of other trophic levels could disentangle how trophic structure is influenced by isolation. An impact of fragmentation on the contraction of trophic structure would be consistent with other studies. Layman et al. (2007a) concluded that fragmentati on simplified trophic struct ure of a tidal creek food web. That simplification was evident in intraspecific comparisons of isotopic ratios from generalist consumers (gray snapper ( Lutjanus griseus Linneaus)) in frag mented and intact tidal creeks. Smallscale experimental studies have also found that fragmenta tion affects trophic structure. In a microbial food web, for example, fragmentation reduced the density of a top predator (Holyoak 2000) In another experimental study, Gilbert et al. (1998) f ound that fewer predator species persisted in moss patches that were isolated than in patches tha t were connected by corridors. Our findings are congruent and extend these results to a much larger spatial scale. Although our results suggest that corridor s may affect trophic position of fire ants via differences in plant species richness, we note that the strong positive relationship between plant species richness and mean 15N of fire ants occurred in just two of three
27 patch types, connected and rectangul ar. We have no explanation for the absence of the relationship in winged patches, but it suggests that increased prevalence of edges in isolated patches may disrupt the relationship between the plant richness and trophic position of fire ants and certainly calls for additional research. The pattern in connected and rectangular patches, however, coincides with studies documenting positive relationships between plant diversity and consumer abundance and diversity (Siemann et al. 1998, Haddad et al. 2009) Fur thermore, experiments by Haddad et al. (2009 and 2011b) found that plant species richness is positively related with species richness of consumers and with abundance of arthropod predators. In conclusion, we teased apart potential effects of patch shape and connectivity on the isotopic signature and likely trophic position of a generalist consumer, and found that connectivity effects have the most influence. These results have relevance in the context of land management and conservation because they sugge st that habitat corridors can help maintain food web structure in fragmented landscapes.
28 Figure 2 1 Corridor experimental landscapes at Savannah River Site, SC. A) Photograph of one of ten experimental landscapes taken from a center patc h looking down the corridor to the connected patch (p hoto courtesy of Julian Resasco) B) Sampling design within an experimental landscape. Dotted lines depict a 3x3 grid centered in each patch, where S. invicta workers were collected in each of the 33.3 m2 cells.
29 Figure 2 2 Corridor and patch shape effects on fire ant mean and range of 15N. A) Corridor effects on mean 15N. B) P atch shape effects on mean 15N. C) C orridor effects on range of 15N D) P atch shape effects on range of 15N For A C, n = 10. For D, n = 9. Bars represent mean SE.
30 Figure 2 3 R elationship between species richness of plants and mean 15N of fire ants. A) In connected patches, B) rectangular, and C) winged patches
31 Figure 24 Relationship between patch fire ant abundance and mean fire ants 15N. Open circles represent patches from polygyne blocks (see Chapter 3) and closed circles represent patches from monogyne blocks.
32 CHAPTER 3 LANDSCAPE CORRIDORS CAN INCREASE INVASION BY AN EXOTIC SPECIES AND REDUCE DIVERSITY OF NATIVE SPECIES Although evidence for positive effects of landscape corridors strips of habitat that connect otherwise isolated habitat patches has amassed from many studies (Tewksbury et al. 2002, Damschen et al. 2006, Gilbert Norton et al. 2010), concerns remain about potential negative effects (Simberloff and Cox 1987, Simberloff et al. 1992, Orrock and Damschen 2005, Weldon 2006). In particular, corridors may facilitate the spread of invasi ve species, which commonly threaten biodiversity and disrupt ecological processes (Simberloff and Cox 1987, Wilcove et al. 1998, Mack et al. 2000, Proches et al. 2005). This potential drawback of corridors, although not documented (Damschen et al. 2006), is critical to evaluate because the same principles that support corridor establishment for threatened species in fragmented landscapes suggest that corridors can simultaneously jeopardize entire communities through spread of invasive species. We use differe nces in dispersal behavior between the two social forms of an invasive ant to predict their use of corridors and examine subsequent im pacts on native ant diversity. Our approach draws on recent work suggesting that species traits and behaviors related to m ovement are important for predicting species response to altered landscape configuration (Damschen et al. 2006, Damschen et al. 2008, Minor et al. 2009, Sullivan et al. 2011). Variation in traits within species are often ignored in ecological studies but c an have profound implications (Crutsinger et al. 2006, Bolnick et al. 2011). H ere we focus on how variation within a species (dispersal behavior of two social forms) affects their response to corridors
33 Ants are model organisms for studying impacts of cor ridors because they are diverse, ecologically important, abundant, and include some of the most widespread and damaging invasive species (Moller 2006). We focused on Solenopsis invicta (henceforth, fire ant) because they are responsible for displacement of native ants and a wide variety of other taxa (reviewed in Holway et al. 2002 and Tschinkel 2006, but see King and Tschinkel 2006 and King and Tschinkel 2008) they commonly occur in open, early successional habitats (Tschinkel 2006; Porter and Savignano 1990), and they also commonly use linear open, early successional habitats (e.g., poweline cuts, roads) that could serve as corridors to spread into new areas (Stiles and Jones 1998). Fire ants have two genetically and ecologically distinct social forms, monogyne and polygyne, defined by the number of reproductive queens per colony (Ross and Keller 1998); monogyne colonies contain a single egg laying queen per nest and polygyne colonies contain multiple egg laying queens per nest. The two social forms also differ in two ways highly relevant to conservation and management: First, ecological impacts of the polygyne social form are more severe than those of their monogyne counterparts. In particular, polygyne invasions result in very high population densities (Macom and Porter 1996) that can devastate native ant communities, whereas monogyne invasions are comparatively more benign (Tschinkel 2006, King and Tschinkel 2006, Porter and Savignano 1990). Second, monogyne queens participate in mating flights at heigh ts > 100 m, dispersing up to several kilometers where they establish new, spatially independent colonies (Markin et al. 1971). In contrast, polygyne queens typically disperse by ground or in low (~ 2 m), short distance flights, establishing new colonies wi thin several meters of their native colony (Tschinkel 2006, Vargo and
34 Porter 1989, Tschinkel 1998). This fundamental difference in queen dispersal behavior suggests the two social forms will respond differently to habitat corridors, with the more dispersal limited and damaging polygyne social form benefiting from corridors the most Given how the two social forms differentially impact native species, the overall balance of negative and positive effects of corridors on ant communities may hinge on the movement ecology of the two social forms of fire ants. Using a landscapescale randomized block experiment with open, early successional habitat patches either connected by a 150m corridor or unconnected, but otherwise equiv alent in area and shape [Figure 3 1; comparisons of connected and unconnected patches of equivalent area but different shape and are also included below], we tested effects of corridors on abundance of fire ants (both social forms) and on the diversity of native ants. Specifically, we tested the predictions that corridors promote invasion of polygyne fire ants by alleviating dispersal limitation, and that their increased dispersal negatively impacts native ant diversity in connected patches. Methods Study a rea We conducted our research at th e Savannah River Site, South Carolina, USA (33.20N, 81.40W) in eight experimental landscapes (blocks), designed to examine effects of corridors and patch shape on movement of plants and animals (Tewksbury et al. 2002). The blocks were created in 1999200 0 (n = 6) and 2007 (n = 2). Each block contained an array of patches of early successional habitat undergoing restoration to longleaf pine savanna embedded in a forested matrix of mostly loblolly pine ( Pinus taeda ) and scattered hardwoods (Fig. 31). Longl eaf pine savanna is an endangered habitat of high conservation interest (Noss 1989, Van Lear et al. 2005,
35 Jose et al. 2006) and high ant diversity (Lubertazzi and Tschinkel 2003). All blocks were assisted in restoration toward longleaf pine savanna by pres cribed burns every 23 years, removal of hardwoods, and planting of native species in an equal fashion across patch types and blocks (details in Tewksbury et al. 2002). Each block consists of a square, central patch (100 x 100 m) surrounded by four peripheral patches of the same habitat (Fig. 31). Each peripheral patch is 150 m from the central patch and is one of three, randomly assigned patch types: connected, rectangular, and winged. The connected patch is 100 x 100 m, with a 150 x 25 m cor ridor that connects to the center patch (1.375 ha total). Winged patches are also 100 x 100 m but have two 75 x 25 m deadend corridors (wings) extending from opposite sides. Rectangular patches are 100 x 137.5 m; the additional area (100 x 37.5m) that m akes these patches rectangular is equivalent to the area of the corridor of connected patches or to the wings of the winged patches. Thus, all patches have the same total area. Each block has one duplicate winged or rectangular patch (Fig. 31). In this st udy, we used only one, randomly selected duplicate patch from each block. By comparing response variables from connected and winged (unconnected) patches, this design allows us to test for effects of corridors while controlling for patch shape and area. An t Sampling. Within connected, rectangular, and winged patches, we sampled ants using pitfall traps, a standard method for measuring abundance and species composition of ground dwelling ants (Bestelmeyer et al. 2000) that has been effective for sampling art hropods in a previous study at our sites (Orrock et al. 2011). We deployed 12 pitfall traps per patch. Traps were placed along 50 m transects that extended diagonally from the four corners of a given patch into the patch center (Fig. 3-
36 1). Traps were place d along these transects at 0 m, 21.5 m, and 50 m. Each trap consisted of a 15dram (28.6 mm inner diameter) plastic vial, onethird full of 50% propylene glycol, inserted flush with the soil surface. To reduce digging in effects, we left traps capped for 48 hours (Greenslade 1973). Once uncapped, traps were open for 48 hours. All ants were counted and identified to species, except for ants in the Aphaenogaster rudis species complex and the Solenopsis (Diplorhoptrum) molesta group, both of which are morpholog ically similar Voucher specimens were deposited in the California Academy of Sciences (http://www.antweb.org). As a second measure of fire and abundance, we used data on the presence of active, mature fire ant nests in one of each patch type per block. One of each duplicate patch was randomly selected to be surveyed in each block. We collected these data by systematically walking and inspecting the entire patch area. Nest counts provide a good metric of density because nests (rather than individuals) ar e the reproductive units of social insects (Gotelli et al. 2011). However, because fire ant nests are often cryptic, we used pitfall incidence as our primary measure to quantify abundance. Social form determination We collected approximately 30 workers from four fire ant nests per patch, one from each quarter of each patch, in 2008 and 2009 and stored samples in 95% alcohol pending DNA analyses. DNA extractions were performed on pooled samples of 1015 workers per nest. All extractions were done using t he Puregene DNA extraction kit (Gentra Systems Inc., USA) following the suggested protocol for extracting DNA from animal tissues. Bulk extracted DNA samples were used as the template for determination of social form, first using a diagnostic Gp9 polymerase chain reaction (PCR) assay developed by Valles and
37 Porter (2003) and then using a second series of PCR assays that more reliably distinguish variation within the class of alleles associated with the expression of polygyny (b like alleles) not detected by th e first assay (see Shoemaker and Ascunce 2010 and Yang et al. 2012 for details regarding social form determination using the informative gene Gp9). In all cases, one social form dominated each block. In blocks designated as monogyne blocks (n = 5 bloc ks), 99% of samples were homozygous, indicating monogyny. In blocks designated as polygyne blocks (n = 3 blocks), 87% of samples were heterozygous, indicating polygyny; however two patches in these blocks had both social forms present. Assignment of soc ial form to block occurred naturally. We do not know why the polygyne social form established in some blocks and not others. Because fire ant establishment is facilitated by habitat alteration (e.g., soil disturbance, Tschinkel 2006), w e tested whether the polygyne form was more likely to appear in landscapes that were historically (prior to forest planting in 1951) in agriculture or in forest, using aerial photograph covering our sites prior to conversion to pine plantation forest. We found that most patch es had an agricultural history (80% of patches from monogyne blocks and 67% of patches from polygyne blocks had over half of the patch area in agriculture in 1951). We found both social forms in patches with historical forest and agriculture cover. The ass ociation between presence of agricultural history and social form was not statistically significant (Fishers exact test, P = 0.65). Because Stiles and Jones (1998) found that fire ants nests were more abundant along linear habitats oriented east/west, we tested whether polygyne forms were found in blocks with corridors oriented in a consistent direction relative to north. We measured
38 the direction of the corridors and wings in each of our blocks using the Google Earth ruler tool, and transformed these val ues as the absolute value of the cosine of the corridor direction in radians. We found corridor and wing directions did not differ between monogyne blocks and polygyne blocks (twosample t test, P Statistical A nalyses. To determine differences in fire ant pitfall trap incidence and number of worker in pitfall traps between polygyne and monogyne blocks, we used MannWhitney U tests. To determine effects of corridors on abundance of fire ants, we tested for an interaction between block social form and patch type on fire ant pitfall trap incidence, with block as a random effect, using a generalized linear mixed effect model. We specified a binomial distribution, a logit link, and the KenwardRoger method of degrees of freedo m approximations. Comparisons between connected and winged patches provide the most direct test of connectivity effects, as they test for effects of connectivity while controlling for patch shape (Tewksbury et al. 2002). We evaluated overdispersion using t he Pearson residuals generated from our analysis. We found that the overdispersion parameter was close to one ( 0.2) indicating an absence of overdispersion (Littell et al. 2006). For pitfall data, incidence is preferable to counts of individual workers w hen estimating density of ants ( Gotelli et al. 2011 ) because individual ant workers are highly aggregated in space. We used a similar generalized linear mixed effect model for nest counts, using a quasi Poisson distribution to adjust for overdispersion. Analyses were performed using R (R Development Core Team 2011) and SAS. To determine effects of corridors on native ants at the community level, we estimated species richness and evenness [using Hurlberts probability of interspecific
39 encounter; "PIE" (Hur bert 1971)] for connected, rectangular, and winged patches. PIE is the probability that two randomly selected pitfall traps from a patch have two different species and is an index that is practically unbiased by sample size. We estimated these metrics with samplebased rarefaction on incidence data from pitfalls using EcoSim (Gotelli and Entsminger 2001) with 1000 iterations. We used the resulting 95% confidence intervals for each patch type to assess differences among patch types, rarified to the number of samples in the patch type with the fewest samples. We randomly removed one duplicated patch type within each block to equalize sampling area for each patch type. We ran analyses separately for polygyne and monogyne blocks. We calculated a nonparametric e stimate of asymptotic species richness, first order jackknife (Jack1) and corresponding confidence intervals (Jack1 2SE) for each patch type using the vegan package (Oksanen 2010) in R (R Development Core Team 2011). We selected Jack1 as a richness estim ator because it is appropriate for incidence data (Hortal et al. 2006) and is among the most accurate species richness estimators for intermediateto high sample coverage (Brose 2003); evaluation of our data showed that the estimator was significantly corr elated with observed levels of species richness (r2 = 0.98). Results We collected and identified 10,775 ants of 49 species using pitfall traps (Table 31). Fire ants made up 99.8 % of nonnative ants and 68% of all ants. Genetic assessment of the spatial distribution of fire ant social forms within our eight experimental blocks [each consisting of five habitat patches; (Fig. 31)] revealed that three blocks were dominated by the polygyne social form (henceforth "polygyne blocks") and five were dominated by the monogyne social form ("monogyne blocks"). Similar to
40 previous studies (Macom and Porter 1996, Vogt et al. 2009), social form had a strong effect on fire ant density. Pitfall traps from polygyne blocks contained over ten times as many individual fire ants and 1.6 times the pitfall trap incidence of fire ants as those from monogyne blocks ( P 's < 0.001). Data on fire ant nest counts further substantiated an increase in fire ant abundance in polygyne blocks (Fig. 32). Corridors significantly increased t he abundance of polygyne but not monogyne fire ants, as reflected by a significant interaction between the dominant social form present in a given block and patch type (with or without corridor) on fire ant pitfall trap incidence (generalized linear mixed model; interaction: F1,12 = 7.48, P = 0.02; patch type: F1,12 = 3.49, P = 0.09; block social form: F1, 6.9, = 5.64, P = 0.05). Corridors had a positive effect on pitfall trap incidence in polygyne blocks, averaging 36% higher pitfall trap incidence compared to unconnected patches (F1,12 = 6.67, P = 0.02; Fig. 3 2 and Fig. 3 3A). In contrast, corridors had no effect on pitfall trap incidence in monogyne blocks (F1,12 = 0.91, P = 0.36) (Fig. 32 and Fig. 33A). Native ant species diversity in polygyne blocks was lower in connected patches than unconnected patches (95% CI in Table 3 2 and Fig. 34). In polygyne blocks, connected patches had 23% lower species richness and 11% lower evenness than unconnected patches (Fig. 33B and Fig. 34, and Table 32). In mo nogyne blocks, however, there was no corridor effect on species richness or evenness (Fig. 33B and Fig. 3 4). The lower species diversity in connected patches of polygyne blocks is likely due to the higher fire ant abundance in those patches (Fig. 33B) Across all patch types, fire ant pitfall incidence was negatively correlated with native ant species richness (Fig. 3-
41 3C; overall: r2 = 0.65, P < 0.001; from monogyne blocks: r2 = 0.56, P = 0.001; from polygyne blocks: r2 = 0.70, P = 0.004). Patches in which fire ants were absent from all pitfall traps had approximately six times as many native species as patches in which fire ants were present in every trap. Discussion Taken together, our results reveal how the net effect of open corridors on ant communities hinges on dispersal and competitive traits associated with the social form of an invasive ant species. T his work highlights the importance of within species in dispersal on connectivity. T he different dispersal behaviors of mon ogyne and polygyne queens determined their response to corridors, and di fferences in colony density of the two social forms in turn determined impacts on native ant communities. Habitat patches connected by corridors are expected to harbor higher, not lower, species richness than unconnected patches of the same area (Gonzalez et al 1998, Damschen et al. 2006, Gilbert Norton et al. 2010). Yet, in patches dominated by polygyne fire ants, we found negative effects of corridors at the community level; species richness and evenness were both lower in connected than in unconnected patches. In agreement with our findings, a study that used previous data from our monogyne blocks found that corridors decreased evenness of ant genera, likely due to the spread of anot her dominant, although native, ant genus, Dorymyrmex (Orrock et al. 2011). By increasing abundance of polygyne fire ants, patch connectivity via corridors negates potential benefits of corridors to other ant species. Roads and powerline cuts are good habit at for fire ants and could serve facilitate the invasion of fire ants into new areas (Stiles and Jones 1998).
42 Effects of fire ant invasion on ant communities during the initial phase of invasion can be dramatic, but may diminish over time (Morrison 2002). This pattern may be the case for invasions in general (Strayer et al. 2006). Indeed, two of the three polygyne blocks were created more recently than the monogyne blocks and only those two blocks showed higher fire ant abundance in connected patches. Futur e work will be needed to investigate the potentially transient nature of invasions and the specific role that corridors play in the long term abundances of both invasive and native species. While concerns about corridors spreading invasive species were fir st raised decades ago (Simberloff and Cox 1987), it is perhaps surprising that empirical examples have not previously been reported. This may be because invasive species often have high dispersal capabilities (Bufford and Daehler 2011), as is the case with the monogyne form of fire ants. Given that species with strong movement capacities are typically not dispersal limited, corridors are unlikely to facilitate their spread; indeed, landscape connectivity appears generally more important for native than for invasive species (Damschen et al. 2006, Minor et al. 2009). Some invasive species, however, have inherently poor dispersal abilities but spread well through accidental human transport (Mack and Lonsdale 2001) [e.g., polygyne fire ants (Tschinkel 2006, King et al. 2009) and Argentine ants, Linepithema humile (Suarez et al. 2001)]. In these cases where invasives are dispersal limited (but successful by other measures), corridors are likely to facilitate spread and could result in a net negative impact on nati ve species. Focusing on species traits will provide guidance on when corridors will spread invasive species across conservation lands.
43 Table 31 Species list and abundances Introduced species are demarcated by an asterisk (*). Subfamily Species Individuals I ncidence Dolichoderinae Dorymyrmex bureni 1134 106 Dorymyrmex grandulus 8 5 Forelius pruinosus 255 67 Lasius neoniger 30 7 Nylanderia arenivaga 11 5 Nylanderia concinna 34 19 Nylanderia faissonensis 11 6 Nylanderia parvula 36 23 Tapinoma sessile 3 1 Ecitoninae Neivamyrmex opacithorax 8 2 Neivamyrmex texanus 3 2 Formicinae Brachymyrmex depilis 20 6 Brachymyrmex patagonicus* 2 2 Camponotus castaneus 9 8 Camponotus chromaiodes 1 1 Camponotus socius 3 2 Formica dolosa 22 12 Formica integra 26 2 Formica pallidefulva 2 2 Formica subsericea 3 2 Myrmicinae Aphaenogaster ashmeadi 10 9 Aphaenogaster carolinensis 9 7 Aphaenogaster floridana 30 22 Aphaenogaster tennesseensis 3 1 Aphaenogaster treatae 10 9 Crematogaster ashmeadi 2 2 Crematogaster lineolata 346 70 Crematogaster missouriensis 4 3 Cyphomyrmex rimosus* 16 8 Monomorium minumum 1 1 Myrmecina americana 6 6 Pheidole crassicornis 296 48 Pheidole dentata 22 13 Pheidole dentigula 4 2 Pheidole metallescens 31 13 Pheidole morrisii 471 28
44 Table 31 Continued. Subfamily Species I ndividuals I ncidence Myrmicinae Pogonomyrmex badius 134 35 Solenopsis geminata 191 7 Solenopsis invicta* 7324 165 Solenopsis molesta grp 193 77 Strumygenys creightoni 4 3 Strumygenys louisianae 3 1 Strumygenys missouriensis 1 1 Strumygenys rohweri 1 1 Strumygenys talpa 2 2 Temnothorax pergandei 18 10 Temnothorax texanus 1 1 Trachymyrmex septentrionalis 12 11 Ponerinae Hypoponera opacior 9 6
45 Table 32 D iversity measures of native ants by patch types Patch type Rarefied species richness 95% Conf. Low 95% Conf. High rarefied to (#samples) Jack1 Jack1 SE Rarefied PIE 95% Conf. Low 95% Conf. High rarefied to (samples) Monogyne blocks (n = 5) Connected 28 25 30 44 43 3.0 0.926 0.920 0.931 44 Winged 30 26 32 44 46 3.0 0.925 0.915 0.934 44 Rectangular 32 28 34 44 52 4.2 0.942 0.937 0.946 44 Polygyne blocks (n = 3) Connected 10 9 11 19 17 2.4 0.807 0.760 0.834 19 Winged 13 11 16 19 26 2.9 0.910 0.888 0.931 19 Rectangular 15 12 18 19 28 3.4 0.907 0.881 0.929 19
46 Figure 31 Aerial photograph (photo courtesy of Ellen Damschen) and layout of one block (n = 8 blocks). Shaded patches (connected, winged, and rectangular) were sampled with pitfall traps. One patch is enlarged to show pitfall trap sites, which were positioned 0m, 21.5m, and 50m from patch corners towards the center.
47 Figure 3 2 Effect of patch type on fire ant abundance, measured as proportion of traps with fire ants and number of nests per patch for patch types. Bars indicate mean s 1 SE bars. The left side shows only monogyne blocks (n = 5) and the right panel shows only polygyne blocks (n = 3). Patch type x block social form interactions were significant for proportion of traps with fire ants and marginally significant for number of nests (pit fall model: interaction: F2, 18 = 5.38, P = 0.01, patch type: F2, 18 = 3.28, P = 0.06, block social form: F1,6.2= 3.33, P = 0.1; nests model: interaction: F2, 11.9= 2.64, P = 0.1, patch type: F2, 11.9= 4.58, P = 0.03; block social form: F1, 6.1 = 0.2, P = 0.67). Within polygyne blocks, patch type effects were significant for both pitfall incidence ( F2, 18 = 5.13, P = 0.02) and nests ( F2, 11.9 = 8.64, P = 0.01). Patch type effects were not significant in monogyne blocks (pitfall: F2, 18 = 0.56, P = 0.6; n ests: F2, 11.9 = 0.35, P = 0.7).
48 Figure 33 Effect of corridors on fire ant abundance and species richness and the relationship between fire ant abundance and species richness. A) Effect of corridors on fire ant abundance, measured as proportion of tr aps with fire ants in connected and unconnected patches (unconnected patches are represented by winged patches, which are equal in area and nearly equivalent in edge:area ratio to connected patches; error bars indicate 1SE). Blue bars indicate monogyne blocks (n = 5) and orange bars indicate polygyne blocks (n = 3). Asterisks indicate statis tical significance (P < 0.05). B) Effect of corridors on native ant species richness, rarefied to lowest number of samples per patch type. Error bars indi cate 95% conf idence intervals. C) Negative correlation of fire ant abundance and species richness of native ants in both monogyne (blue circles) and polygyne blocks (orange squares)
49 Fig ure 3 4 Species richness and evenness ( Hurlberts P IE) sample based rarefaction curves for patch types. A) Species richness and B) Hurlberts PIE for polygyne blocks (n = 3) C) Species richness and D) Hurlberts PIE for monogyne blocks (n = 5) C onnected patches are in red, rectangular are in black, and winged are in blue. Dashed lines indicate upper and lower bounds of 95% confidence intervals
50 CHAPTER 4 TESTING SODIUM LIMIT ATION OF FIRE ANTS IN THE FIELD AND LABO RATORY Sodium is vital for the physiological functioning of animals (Frausto da Silva & Williams 2001) but is often r are in the environment (Kaspari et al. 2008; 2009). Because sodium deposition varies geographically and is largely determined by distance from the coast, coastal ecosystems typically have higher sodium deposition than do inland ecosystems (National Atmosph eric Deposition Program 2011). G eographic variation in sodium deposition influences the behavior of organisms that rely on sodium for functioning. For instance, preferential foraging for sodium by herbivorous and omnivorous ants tracks sodium availability in the environment; thus, coastal ants respond less strongly to sodium baits than do inland ants (Kaspari et al. 2008). Carnivorous ants, which presumably receive sufficient sodium through their prey, do not respond strongly to salt baits at any location t ested (Kaspari et al. 2008). The implications of changes in sodium deposition by altered patterns of precipitation due to climate change have been suggested to have profound consequences on the carbon cycle by ant facilitated increases in decomposition in inland carbon pools like the Amazon (Kaspari et al. 2008; 2009; Dudley et al. 2012). However, these implications are only based on behavioral responses of ants to salt baits. W hile preferential foraging for sodium by ants is extensively documented (Vail et al. 1999; Kaspari et al. 2008; 2010; ODonnell et al. 2010; Arcila Hernndez et al. 2012; Chavarria Pizarro et al. 2012), the effect of increased sodium supplementation on for colony growth is untested. Red imported fire ants, Solenopsis invicta Buren (hereafter fire ants), are an invasive species that that is widely established in the southeastern United States and several regions throughout the world (Tschinkel 2006). They are a numerically dominant
51 species in disturbed areas in their introduced range. If altered patterns of sodium deposition indeed affect ant abundances, then important ecological and economic impacts of increased fire ant abundance are likely. In addition, fire ants have a flexible trophic position (Tillberg et al. 2006; Resasco et al 2012) and therefore it is uncertain whether they can meet their sodium demands solely from animal prey. In this study, we aim to determine if there is a link between behavioral responses to salt and colony growth. To do this we (1) conducted bait trials to determine whether fire ants show stronger preferences for high concentrations of NaCl in inland sites than costal sites and (2) reared colonies of fire ants in the laboratory to test whether supplemental NaCl at different concentrations increased colony growth. Methods Sodium bait foraging e xperiment We conducted NaCl bait trials in August and September 2012 in two regions of the southeastern United States that differ in NaCl deposition rate (National Atmospheric Deposition Program 2011). The Savannah R iver Site, a National Environmental Research Park, South Carolina (henceforth SC) receives approximately 1.5 kg ha1 yr1 Na+, whereas St. Johns Co. in northeast Florida 1 yr1 Na+. At six sites within each region, we co nducted preference trials modified from the protocol of Kaspari et al. (2008). At 1m intervals along one 100m transect per site, we placed randomly drawn 1.5mL plastic, uncapped centrifuge tubes flat on the ground. All tubes were half filled with cotton wads saturated with either a sucrose solution (10% sucrose), deionized (DI) water, or a NaCl solution with one of three concentrations (0.1%, 1%, and 2% NaCl). Sucrose solutions were used as a measure of ant activity independent of geographic region so that responses to NaCl baits could be scaled by responses to sucrose baits (Kaspari et al.
52 2008). Thus, all comparisons of fire ant responses among NaCl baits are scaled by sitespecific responses to sucrose. Each bait type was replicated 20 times within eac h transect for a total of 100 baits per transect. After one hour, we picked up the tubes and quickly closed the caps to trap the ants within. We used a two way ANOVA with Type III sum of squares to assess the interaction between geographic region (SC and FL) and NaCl bait concentration (0%, 0.1%, 1%, or 2%) on fire ant recruitment to field baits. The response variable was the proportion of each NaCl bait concentration occupied by fire ants, divided by the proportion of sucrose baits occupied at each site. Two sites in FL were not used in the analysis because of lack of fire ant activity at baits. We checked for normality on Q Q plots and tested homoscedasticity with Bartletts test. A square root transformation on the response variable improved heteroscedas ticity. We used a Tukeys HSD to make pairwise comparisons among combinations of geographic region and NaCl bait concentrations. Colony growth e xperiment On May 31, 2012 we collected approximately 150 mated queens in SC after a large nuptial flight. During the claustral period, we placed queens into individual test tubes with access to DI water. We discarded tubes with queens that had died, had sexual brood, or very limited worker brood (all signs of various pathologies). Seven weeks later, we categorized colonies by the number of workers and brood as either large (274.0 SD workers colony1) or small (18.24.05 SD workers colony1). Within each size class w e randomly assigned colonies to one of four treatments -access to 0% NaCl (DI water), 0.01% NaCl, 0.1% NaCl, or 1% NaCl --
53 while maintaining equal numbers of each size class across treatments. Each treatment received seven replicates of large colonies an d eight replicates of small colonies. Each colony was kept in a fluonlined, opentop plastic container (Fig. 41). Rigorous hygiene procedures were used to eliminate contamination with SINV 3 and other pathogens (Valles and Porter 2013). Colonies were randomly assigned to plastic bins with lids that held up to ten containers. The bottoms of these bins were dusted with talcum powder to prevent cross contamination in case ants escaped. All treatments received: (1) ad libtum sucrose in wicks moistened three times a week with DI water and replaced as often as needed, (2) ad libitum chopped superworms ( Zophobas morio ; reared on a diet of carrots, potatoes and wheat middlings; Timberline Live Pet Food, Marion, IL, USA), (3) ad libtum supplemental NaCl solution or DI water (treatments) administered in test tubes plugged with cotton, and (4) nesting tubes filled with DI water. We selected superworms because they are relatively low in sodium compared to a common alternative, domestic crickets ( Acheta domesticus ; 0 .33 mg g 1 vs 0.97 mg g1, respectively; J. Resasco, unpubl. D ata see below ), and because fire ant colonies cannot be reared on vegetable or artificial diets (Porter 1989; S.D. Porter unpublished data). NaCl solutions were replaced every 23 weeks befor e more than half of the water had evaporated to minimize the effect of altered concentrations. During the course of the experiment we eliminated eight colonies with dead queens, high worker mortality, or very low brood numbers. After eleven weeks, we measured fresh biomass of entire colonies (queen, workers, and brood) to the nearest 0.01 g. We used a two way ANOVA to test the effect of NaCl supplementation on fire ant colony mass. Initial colony size (small or large) and NaCl concentration (0%, 0.01%,
54 0. 1%, and 1%) were factors. Again, we checked for normality on Q Q plots and tested homoscedasticity with a Bartletts test. No data transformation was required. At the conclusion of the colony growth experiment, we sampled 3050 workers from each colony an d combined them by treatment. Sodium content analysis was performed at ABC Research Laboratories, Gainesville, FL, USA ( www.abcr.com ), using method AOAC 985.01. Results We found a highly significant interaction between geographic region and NaCl bait concentration on the response of fire ants (Fig. 42 ; F3,32 = 13.36, P << 0.0001). At low concentrations (0% and 0.1% NaCl), fire ant recruitment to baits did not differ between sites ( P < 0.99), but at high concentrations (1% and 2% NaCl), fire ants in SC showed a strong preference for NaCl baits. Response to high concentration baits in SC was roughly an order of magnitude greater than for the same baits in FL ( P < 0.01) and for lower conc entration baits (0% and 0.1%) at both sites ( P < 0.001). In FL, response did not differ among concentrations ( P Despite the differences in response to NaCl in the field, we found that NaCl supplementation had no effect on colony growth ( F3, 44 = 0.25, P = 0.86) (Fig. 4 3 ). Analysis of workers after the conclusion of the experiment revealed that sodium content in the lab colonies did not increase with increasing NaCl treatment concentration (Fig. 44 ). These results support the expectation that the response of ants to sodium is inversely related to geographic differences in its rate of deposition (Kaspari et al. 2008). These results also suggest that at least inland fire ant populations do not meet their sodium needs from animal prey.
55 Discussion These results suggest that our experimental colonies were not sodium limited. Perhaps they were able to fulfill their sodium requirement through consumption of superworms. Alternatively, they may have ingested their own fecal material as a concentrated so urce of sodium (M. Kaspari personal communication). Regardless, our results show that increasing sodium availability provided no benefits to colony growth. In agreement with our findings, experimentally added dietary sodium does not lead to increases in ei ther survivorship or reproduction in Japanese beetles (Stamp and Harmon 1991). Future work should be directed at rearing colonies on sodium free and sodium supplemented diets to better isolate any growth benefits of sodium or by manipulating sodium availability in the field Such work will help assess the ecological implications of altered sodium deposition on inland ants and the carbon cycle.
56 Figure 4 1 Photograph of one experimental fire ant colony (p hoto courtesy of Julian Resasco)
57 Figure 4 2 Fire ant response to NaCl baits in Florida and South Carolina. Mean ( 1 SE) of ratio of fire ant activity at NaCl baits of increasing concentration (0% [DI water], 0.1%, 1%, and 2%) to 10% sucrose bait activity. High sodium deposition sites in Florida (n = 4) are depicted in red circles and a solid line, while low sodium deposition sites in South Carolina (n = 6) are depicted in with blue triangles with a dashed line.
58 Figure 4 3 NaCl treatment effects on experimental fire ant colonies. Mean colony mass ( 1 SE) of laboratory reared, experimental fire ant colonies with salt solution supplements of 0% NaCl (DI water; n = 14 colonies), 0.01 NaCl (n = 14), 0.1% NaCl (n = 13), and 1% NaCl (n = 11).
59 Fig ure 4 4 Sodium content of combined samples of fire ants from each of the four NaCl supplementation treatments. Supplementation treatments were 0% NaCl (DI water), 0.01 NaCl, 0.1 NaCl, and 1% NaCl.
60 CHAPTER 5 USING HISTORICAL AND EXPERIMENTAL DATA TO REVEAL WARMING EFFEC TS ON ANT COMMUNITIES Global climate change has been documented to alter phenology and species ranges, and to cause extinction (reviewed in Parmesan 2006). Predicting how communities will change under a changing climate is a pressing challenge for ecologists. Observational studies that examine the relationship between climate trends or weather events and changes in biotic communities have a long history in ecology (Brown et al. 1997). In cases where historical data exist, repeating sampling protocols and comparing historical and contemporary datasets can reveal community changes that have occurred under decades of climatic change (Sagarin et al 1999; Smith et al 2006; Damschen et al. 2010). In essentially all cases, however, confounding factors (e.g., succession, poll ution, changes in soil, invasion, landscape context) make it difficult to attribute observed differences to changes in climate (Parmesan 2006). Field experiments that sim ulate projected climatic change on organisms can provide a bridge between long term correlative studies and the potential mechanisms that underlie any observed patterns. T hey increase ability to assign causation of biotic changes to abiotic variables. Howe ver experimental studies have other limitations such as limited replication, and small spatial and temporal scales. Short temporal scales may often not capture the relevant important climatic changes (e.g., e xtreme weather events, variance) and biotic changes that are slow to emerge. A combined approach of long term observations and experimental manipulation can overcome many of the inherent limitations of detection and attribution of such changes. Here we compare communities of ants under the environmental warming of the last few decades based on historical surveys and interpret the results based on an
61 artificial warming experiment. Ants are diverse, abundant, and ecologically important (Chapter 1 and Hlldobler and Wilson 1990). Temperature is a key component of climate change for ants to investigate because it : correlates with patterns of ant species diversity and abundance (Kaspari et al. 2003, Sanders et al. 2007, Dunn et al. 2009) affects the timing of ant reproduction (Dunn et al. 2007) and foraging behavior ( Porter and Tschinkel 1987, Ruano et al. 2000), limits species ranges ( Korzukhin et al. 2000, Morrison et al. 2005, Diamond et al. 2012), and affects ant colony growth and development (Porter 1998). M ethods Study s ystems We conducted this study at two sites approximately 450 km apart: Savannah River Site (SRS), South Carolina, (33.208 N, 81.408 W) and Duke Forest, North Carolina (35.520 N, 79.590 W). At SRS, our sampling area was in turkey oak ( Quercus laevis ) forests. At Duke Forest, our experimental site was located in an oak hickory forest. Both sites have been undisturbed for the last 80 years. Historic d ata: Savannah River Site. Data on abundance and diversity of ants at SRS were collected in the summer of 1976 by V an Pelt and Gentry (1985) and by us in the summers of 2010 and 2011, using the same sites and similar methodology. Van Pelt and Gentry 1985 used 148 mL pitfall traps baited with either sugar or peanut butter solutions. They also used baited containers and hand collections but because sampling effort and abundances for these were not reported we only used thei r pitfall trap data ( Van Pelt and Gentry 1985). We collected using 55mL pitfall traps baited with either sugar or peanut butter solutions. In all collection periods pitfall traps were left open for 24 hours. We sorted and identified ants to species, except for two taxonomically difficult groups ( Solenopsis molesta group and Aphaenogaster rudis complex) in which
62 individuals were combined. To compare the t wo datasets we used current synonyms for species, based on the taxonomic history provided in AntWeb ( www.antweb.org ). To assess the extent of climatic warming between the historic and present day sampling periods, we obtained data on mean temperatures between 19762011 from a nearby weather station, Aiken, SC 5SE (33.49N, 81.70W) SC State Climatology Office http://www.dnr.sc.gov/climate/sco/). Missing data (~12% of months) were filled in using data from the secondnearest weather station, Augusta, Bush Field (KAGS), GA ( 33.38N 81.97W). Mean summer temperatures (June, July, and August) were approximately 2.7C warmer in 20102011 than in 1976. Over the intervening years mean annual temperatures also show an increasing trend Fig. 5 1 Duke Forest warming e xperiment The Duke Forest warming experiment (35 consists of 12 octogonal opentop chambers centered on > 20 cm oak ( Quercus ) trees. Chambers are constructed of wooden frames walls covered in greenhouse sheeting. Chambers are 5 meter diameter, 1.5 m high (~ 22 m2), and have a 3 cm gap along the bottom of each chamber. These chambers have been actively heated with warm air since January 2001 and are held at ambient temperatures or are continuously heated to one of the 0.5 C intervals between 1.5 to 5.5C above ambient temperature. Temperature is monitored within each chamber. Experimental design is discussed in detail in Pelini et al. 2011 a Within these chambers, we collected data on abundance and diversity of ants using four pitfall traps (90 mL) one third filled with propylene glycol. Pitfall traps were left open for a 48h sampling period. Ants were collected from pi tfall traps and identified to
63 species. Voucher specimens are deposited at North Carolina State University and at Harvard Forest. To correspond with the same season as SRS data, only data from June, July, and August pitfall samples were used. Statistical a n alysis To estimate differences in species richness and evenness between sampling periods at SRS we used samplebased rarefaction on incidence data from 2010 and 2011 pitfall trap data using EcoSim (version 7.71; Gotelli and Entsminger 2001) with 1000 iter ations. We used calculated 95% confidence intervals to compare species richness and evenness between the 1976 data and the estimated present day metrics rarefied to an equivalent sample number. We calculated species turnover using Bray Curtis distance on r elative abundances of ant species among sampling periods of SRS and among temperature treatments at Duke Forest. We used linear regression to examine the relationship between temperature and species richness, evenness, and species relative abundance for both SRS and Duke Forest for species that occurr ed at both sites. We also used M antel tests to examine the relationship between temperature differences and species turnover at both sites. Results Across sampling periods and study sites there were 57 species of ants (Table 51). Only four species occurred across sampling periods at SRS and Duke Forests. However 67% of the species that occurred at Duke Forest are present in the regional species pool at SRS (Chapter 3 and J. Resasco unpublished data). S pecies richness decreased by approximately 35% at SRS between 1976 and 20102011. This difference is outside of the present day 95% confidence intervals constructed by rarifying t o equivalent sample sizes (Fig. 52). However, relatively short -
64 term experimental warming at Duke Forest showed no effect on species richness ( P = 0.99, r2 < 0.001). Evenness decreased by approximately 10% at SRS between 1976 and 20102011. This difference was outside of the 95% confidence intervals constructed by rarifying t o equivalent sample sizes (Fig 52 ). As with species richness, differences in evenness were not apparent in the experimental warm ing treatments at Duke Forest ( P = 0.62, r2 = 0.03). Bray Curtis distance was positively related to mean summer temperature dif ferences at SRS (Fig. 53 ), although the relationship was not statistically significant (Mantel r = 0.90; P = 0.33). Likewise, at Duke Forest, we found that as temperature differences among warming chambers increased, Bray Curtis distance increased, although the relationship was marginally significant (Mantel r = 0.22; P = 0.06). The relative abundance of Crematogaster lineolata more than doubled between the 1976 sampling period and the present day sampling period at SRS. Similarly, there was a positive r elationship between warming and C. lineolata relative abundance in the experimental chambers at Duke Forest, although the relationship was marginally significant ( P = 0.08; r2 = 0.27). The relative abundance of Myrmecina americana and Solenopsis molesta g rp. decreased between sampling period at SRS. In Duke Forest, M. americana also decreased ( P = 0.03; r2 = 0.39) but S. molesta did not show a relationship with temperature ( P = 0.40; r2 = 0.12) Temnothorax pergandei were rare in both sites and did not show a temperature effect ( P = 0.06; r2 = 0.99).
65 Discussion While we found declines in richness and evenness with climate warming, we did not find them for experimental warming. These declines are likely due to successional maturation of forest stands at SRS. The presence of certain ant species like Dorymyrmex sp., Forelius pruinosus Nylanderia arenivaga, Pheidole davisi Pheidole metallescens Phidole crassicornis Pogonomyrmex badius, and Trachymyrmex septentrionalis in the 197677 data but not the present day data suggest that the sampling sites were likely more open and xeric during the original sampling period. Succession can result in large changes in animal communities (Smith 1928, Robinson 1991, Dunn 2004). Other sources of differences could be differences in trap diameters between the sampling periods may result in differences in capture efficiency of rare ants (Abe nsperg Traun and Steven 1995). R easons for community level difference effects between SRS and Duke forest could be due to the entry of ants from outside of the experiment (Moise and Henry 2010) or the comparatively short time scale of the experiment. At the species level, we observed similar responses to warming on some cooccurring species, in particular, a numerically dominant spec ies C. lineolata Under both historic climate warming and experimental warming, C. lineolata increased its representation in the community. An earlier study that used passively warmed chambers at Duke Forest also found C. lineolata abundance increased with increasing temperature while evenness decreased with temperature (Pelini et al. 2011b ). A likely explanation is that C. lineolata benefits by increasing foraging and thus reduces community evenness by competitive displacement. Similarly, an experiential s tudy on plants showed that dominance patterns are shifted under experimentally altered climate.
66 A dominant species, Lespedeza, became more abundant with increased precipitation, which caused a decrease in evenness (Kardol et al. 2010). Walker et al. 2006 l ikewise found that with experimentally increased temperature, woody plants became dominant and species richness and evenness declined. Why do some species benefit from warming while others do not? In the Duke Forest warming experiment, Stuble et al. 2013 f ound that C. linolata and increased recruitment to baits with increased temperature while other species did not. Ant thermal tolerance may be a promising predictor of responses to warming temperatures (Diamond et al. 2012, Stuble et al 2013). In conclusion we found confounding drivers of chage at the community level, succession and warming. However, similarities in responses to warming at the species level in species coocurring ants at SRS and Duke Forest suggest that species are responding to w arming temperature and that there are differential responses among species. Recent work suggests species traits that could be used predict how species will respond to changing temperatures. This study also highlights the importance of assessing alternative explanations and drawing on experimental data to make better inferences about climate change from historical datasets.
67 Table 5 1. Species list for Savannah River Site and Duke Forest. Darkened cells indicate species occurrence. Species SRS 1 976 SRS 20 1 0 11 Duke forest Amblyopone pallipes X Aphaenogaster ashmeadi X X Aphaenogaster fulva X Aphaenogaster lamellidens X Aphaenogaster mariae X Aphaenogaster rudis comp. X X Aphaenogaster tennesseensis X Aphaenogaster treatae X X Camponotus americanus X Camponotus castaneus X Camponotus chromaiodes X Camponotus nearcticus X Camponotus pennsylvanicus X X Camponotus socius X Crematogaster ashmeadi X Crematogaster lineolata X X X Crematogaster minutissima X X Crematogaster vermiculata X Dorymyrmex insana X Forelius pruinosus X Formica dolosa X X Formica pallidefulva X Formica pallidefulva X Formica sanguinea group X Formica subsericea X Hypoponera opacior X Lasius interjectus X Myrmecina americana X X X Myrmecina cryptica X Neivamyrmex texanus X Nylandaria faissonensis X X Nylanderia arenivaga X Nylanderia parvula X X Nylanderia concinna X Nylanderia terricola X Pheidole davisi X
68 Table 5 1. C ontinued. Species SRS 1 976 SRS 20 1 0 11 Duke forest Pheidole dentata X X Pheidole dentigula X Pheidole metallescens X Pheidole morrisi X X Phidole crassicornis X Pogonomyrmex badius X Ponera pennsylvanica X Prenolepis imparis X Pseudomyrmex ejectus X Pyramica bunki X Pyramica carolinensis X Pyramica ornata X X Pyramica pergandei X Pyramica sp DFmorph X X Solenopsis molesta grp. X X X Stenamma cf impar X Stenamma impar X Temnothorax pergandei X X X Temnothorax schaumii X Temnothorax curvispinosus X Trachymyrmex septentrionalis X
69 Fig ure 5 1 Annual, summer, and winter mean monthly temperatures near Savannah River Site, SC 1975 2011.
70 Figure 5 2 Relationships between tem perature and species richness, evenness, and turnover at Savannah River Site and Duke Forest. A B) species richness, evenness C D), and turnover E F ) from historical resampling of ant commu nities at Savannah River Site ( A, C, E ) and experimental warming chambers at Duke Forest (B, D, F ).
71 Figure 5 3 Relationship between temperature and species relative abundances for ants species that occurred at both Savannah River Site and Duke Forest. Species are: A B ) Crematogaster lineolata, C D ) Myrmecina americana, E F ) Solenopsis molesta grp GH ) Temnothorax pergandei
72 CHAPTER 6 SUMMARY AND CONCLUSIONS Chapter 1 Habitat loss and fragmentation invasive species, and climate change are the major threats to biodiversity. In this introductory chapter, I discussed these threats and outlined the experiments that follow in Chapters 2 5. Chapter 2 Habitat fragmentation disrupts species movement, leading to local extinctions and altered community structure. Habitat corridors, which connect isolated patches of habitat and facilitate movement between patches, provide a potential solution to these negative impacts. However, most studies to date have examined the movement of species alone without considering emergent effects on the community (e.g., altered trophic structure). We use lar ge scale, experimental landscapes and 15N) of a common generalist consumer (the fire ant, Solenopsis invicta) to determine how corridors affect trophic structure. Thus, because the fire ant is a species whose trophic posit ion is flexible and whose diet typically reflects local prey availability, we assume that shifts in fire ants trophic position between connected and isolated patches are likely to reflect shifts in patch trophic structure. We found that colonies in isolat ed patches had lower means and ranges of 15N than colonies in otherwise similar connected patches, suggesting that corridors may increase fire ants trophic position and breadth, respectively. Previous work in our landscapes documented higher species rich ness of plants in connected than 15N were positively correlated with plant richness, suggesting that increased plant richness may influence the observed 15N. Together these results suggest that fragmentation may reduce trophic position and narrow trophic breadth of dietary generalists such as the fire
73 ant. These shifts likely reflect an alteration of food webs in isolated patches. Our results suggest that corridors may be effective in preventing or reducing such alterations. Chapter 3 Although corridors are often used to mitigate negative effects of habitat fragmentation, concerns persist that they may facilitate the spread of invasive species. In a largescale experiment, we measured effects o f open corridors on the invasive fire ant, Solenopsis invicta and on communities of native ants. Fire ants have two social forms: polygyne, which disperse poorly but establish at high densities, and monogyne, which disperse well but establish at low densi ties. In experimental landscapes dominated by polygyne fire ants, fire ant abundance was higher and native ant diversity was lower, in habitat patches connected by corridors, compared to unconnected patches. In contrast, in landscapes dominated by monogyne fire ants, habitat connectivity had no influence on fire ant abundance or native ant diversity. Polygyne fire ants dominated recently created habitat patches, suggesting their responses to corridors may be transient. These results demonstrate that corridors can facilitate the invasion of some nonnative species, and highlight the importance of considering species traits when assessing the utility of corridors in conservation. Chapter 4 Sodium is an essential dietary element and preferential foraging for high concentrations of sodium by inland herbivorous and omnivorous ants suggests it may be limiting. If so, increased sodium availability through altered deposition and anthropogenic sources may lead to increased colony growth and cascading ecological impa cts. For red imported fire ants, Solenopsis invicta Buren, we test (1) whether colonies from coastal and inland sites differ in responses to salt baits and (2) whether supplemental NaCl increases growth of fire ant colonies in the laboratory. Fire ants
74 res ponded roughly an order of magnitude more strongly to high concentrations of NaCl baits in inland sites with low sodium deposition than did fire ants in costal sites with high sodium deposition. Laboratory reared fire ants, however, showed no signs of sodi um limitation or benefits of increased sodium. The link between behavioral responses to baits in the field and effects on colony growth, deserve further investigation to assess the ecological impacts of altered sodium availability. Chapter 5 Historical re cords of biotic communities can be compared to current records to suggest effects of recent climate change. However, attribution of biotic changes to climate changes is problematic due to confounding variables. Experiments that manipulate temperature can overcome this issue of attribution, but long term implications of observed community responses from such experiments are uncertain. Here we combine observational and experimental data to understand the effects of warming temperatures on ant communities. Obs ervational data span a 35 year period, during which summer temperatures at the site increased 2.7 C. Experimental data come from an experiment in which temperatures were increased 1.5 5.5 C. Species richness and evenness decreased with warming under natural but not experimental warming. Species turnover, however, tended to increase with temperature in both datasets, suggesting differential effects of warming on different species. Under both experimental and natural warming, abundance of the dominant ant Crematogaster lineolata increased. Myrmecina americana decreased with temperature in both datasets. Solenopsis molesta group did not show a change and Temenothorax pergandei remained rare across temperatures. We conclude that species respond
75 differently to warming temperatures and that community structure of forest ants has changed under recent warming and will continue to change under projected warming.
76 LIST OF REFERENCES Abe nsperg Traun, M., and D. Steven. 1995. The effects of pitfall trap diameter on ant species richness (Hymenoptera: Formicidae) and ant species composition of the catch in a semi arid eucalypt woodland. Australian Journal of Ecology 20: 282 287. Alonso, L E 2000. Ants as indicators of diversity. in : Agosti D J. Majer E. Alonso and T. R. Schultz, e d itors. Ants: standard methods for measuring and monitoring biodiversity. Smithsonian Institution Press, Washington DC USA. Andersen, A. N., B. D. Hoffmann, W. J. Muller, and A. D. Griffiths. 2002. Using ants as bioindicators in land management: simplifying assessment of ant community responses. Journal of Applied Ecology 39:8 17. Arcila Hernndez, L. M., E. V. Todd, G. A. Miller, and M. E. Frederickson. 2012. Salt intake in Amazonian ants: too much of a good thing? Insectes Sociaux 59:425 432. Bearhop, S., C. E. Adams, S. Wa ldron, R. A. Fuller, and H. MacL eod. 2004. Determining trophic niche width: a novel approach using stable isotope analysis. Journal of Animal Ecology 73:10071012. Bestelmeyer B T ., et al. 2000. in : Agosti D J. Ma jer E. Alonso and T. R. Schultz, e d itors. Ants: standard methods for measuring and monitoring biodiversity. Smithsonian Institution Press, Washington DC USA. Blthgen, N., G. Gebauer, and K. Fiedler. 2003. Disentangling a rainforest food web using stable isotopes: dietary diversity in a species rich a nt community. Oecologia 137:426 435. Bolnick, D. I., et al 2011. Why intraspecific trait variation matters i n community ecology. Trends in Ecology and E volution 26:183 92. Brose U., N. D. Martinez, a nd R. J. Williams. 2003. Estimating species richness: Sensitivity to sample coverage and insensitivity to spatial patterns. Ecology 84:2364 2377. Brown, J. H., T. J. Valone, and C. G. Curtin.1997. Reorganization of an arid ecosystem in response to recent c limate change. Proceedings of the National Academy of Sciences ( USA ) 94:9729 9733. Bufford, J. L., and C. C. Daehler. 2011. Life history strategies. in : Simberloff D and M. Rejmnek editors. Encyclopedia of Biological Invasions University of California Press, Los Angeles CA, USA ChavarriaPizarro, L., H. F. McCreery, S. P. Lawson, M. E. Winston, and S. ODonnell. 2012. Sodium specific foraging by leafcutter ant workers ( Atta cephalotes Hymenoptera: Formicidae). Ecological Entomology 37:435 438
77 Collinge, S. K. 2009. Ecology of fragmented landscapes. Johns Hopkins University Press, Baltimore, MD USA. Crutsinger G., M. Collins, and J. Fordyce. 2006. Plant genotypic diversity predicts community structure and governs an Ecosystem Process. Science 647:966 968. Damschen, E. I., L. A. Brudvig, N. M. Haddad, D. J. Levey, J. L. Orrock, and J. J. Tewksbury. 2008. The m ovement ecology and dynamics of plant communities in fragmented landscapes. Proceedings of the National Academy of Sciences ( USA ) 105:19078 19083. Damschen, E. I., N. M. Haddad, J. L. Orrock, J. J. Tewksbury, and D. J. Levey. 2006. Corridors increase plant species richness at large scales. Science 313:1284 128 6. Damschen, E. I., S. Harrison, and J. B. Grace. 2010. Climate change effects on an endemic rich edaphic flora: resurveying Robert H. Whittakers Siskiyou s ites (Oregon, USA). Ecology 91: 3609 3619. Davies, K. F., B. A. Melbourne, and C. R. Margules. 2001. Effects of withinand betweenpatch processes on community dynamics in a fragmentat ion experiment. Ecology 82:1830 1846. Diamond, S. E. L. M. Nichols, N. McCoy, C. Hirsch, S. L. Pelini, N. J. San ders, A. M. Ellison, N. J. Gotelli, and R. R. Dunn. 2012. A physiological trait based approach to predicting the responses of species to experimental climatic warming. Ecology 93:2305 2312. Dudley, R., M. Kaspari, and S. P. Yanoviak 2012. Lust for salt in the western Amazon. Biotropica 44 :6 9. Dunn, R. R. 2004. Recovery of faunal c om munities during tropical forest r egeneration. Conservation Biology 18:302 309. Dunn, R. R., C. R. Parker, M. Geraghty, and N. J. Sanders. 2007. Reprod uctive phenologies in a diverse temperate ant fauna. Ecological Entomology 32:135 142. Dunn, R. R., et al. 2009. Climatic drivers of hemispheric asymmetry in global patterns of ant species richness. Ecology Letters 12:324 333. Ellison, A. 2012. Out of Oz: opportunities and challenges for using ants (Hymenoptera: Formicidae) as biological indicators in northtemperate cold biomes. Myrmecological News 17:105 119. Estes, J. A., et al. 2011. Trophic downgrading of planet earth. Science 333:301 306. Fahrig, L. 2 003. Effects of habitat fragmentation on biodiversity. Annual Review of Ecology Evolution and Systematics 34:487515.
78 Foley, J. A, C. Monfreda, N. Ramankutty, and D. Zaks. 2007. Our share of the planetary pie. Proceedings of the National Academy of Sciences (USA) 104:12585 1258 6. Frausto da Silva, J. J. R. and R. J. P. Williams. 2001. The Biological Chemistry of the Elements: The Inorganic Chemistry of Life. Oxford University Press, Oxford, U.K. Gannes, L. Z., C. Martnez del Rio, and P. Koch. 1998. Natural abundance variations in stable isotopes and their potential uses in animal physiological ecology. Comparative Biochemistry and Physiology Molecular & Integrative Physiology 119 A:725 737. Gibb, H. and S. A. Cunningham. 2011. Habitat contrasts reveal a shift in the trophic position of ant assemblages. J ournal of Animal Ecology 80:119 127. Gilbert, F., A Gonzalez, and I. Evans Freke. 1998. Corridors maintain species richness in the fragmented landscapes of a microecosystem. Pr oceedings of the Royal Soc iety of London Series B 265:577 582. Gilbert Norton, L., R. Wilson, J. R. Stevens, and K. H. Beard. 2010. A metaanalytic review of corridor effectiveness. Conservation Biology 24:660 668. Gonzalez, A., J. H. Lawton, F. S. Gilbert, T. M. Blackburn, and I. Evans Freke. 1998. Metapopulation dynamics, abundance, and distribution in a m icroecosystem. Science 281:2045 2047. Gotelli, N. J. and A. E. Arnett. 2000. Biogeographic effects of red fire ant invasion. Ecology Letters 3:257 261. Gotelli, N J. A. M. Ellison, R. R. Dunn, and N. J. Sanders 2011. Counting ants (Hymenoptera: Formicidae): biodiversity sampling and statistical analysis for myrmecologists. Myrmecological News 15:13 19. Gotelli, N J. and G. L. Entsminger 2001. EcoSim 7.71. (Acquired Intelligence Inc. & Kesey Bear, available at http://homepages.together.net/!gentsmin/ecosim.htm/). Greenslade, P. J. M. 1973. Sampling ants with pitfall traps: Digging in effects. Insectes Sociaux 20:343 353. Haddad, N. M., G. M. Crutsinge r, K. Gross, J. Haarstad, J. M. H. Knops, and D. Tilman. 2009. Plant species loss decreases arthropod diversity and shifts trophic str ucture. Ecology Letters 12:1029 1039. Haddad, N. M., G. M. Crutsinger, K. Gross, J. Haarstad, and D. Tilman. 2011b Plant diversity and the stability of food webs. Ecology Letters 14:42 46. Haddad, N. M., B. Hudgens, E. I. Damschen, D. J. Levey, J. L. Orrock, J. J. Tewksbury, and A. J. Weldon. 2011a. Assessing positive and negative ecological ef fects of corridors. in J. Liu, V. Hull, A. T. Morzillo, and J. A. Wiens, editors. Sources, sinks, and sustainability Cambridge University Press, Cambridge, UK
79 Haddad, N. M. and J. J. Tewksbury. 2005. Low quality habitat corridors as movement conduits for two butterfly species. Ecological Applications 15:250 257. Harrison, S. and E. Bruna. 1999. Habitat fragmentation and largescale conservation: what do we know for sure? Ecography 22:225 232. Hilty, J. A., W. Z. Lidicker, and A. M. Mere nlender. 2006. Corridor ecology : the science and practice of linking landscapes for biodiversity conservati on. Island Press, Washington, DC USA. Holbrook, S. J., R. J. Schmitt, and J. S. Stephens. 1997. Changes in an assemblage of temperate reef fishes associated with a climate shift. Ecolo gical Applications 7 : 1299 1310. Hlldobler B. and E. O. Wilson 1990. The ants. Belknap Press of Harvard University Press, Cambridge, MA, USA Holt, R. D. 1993. Ecology at the mesoscale: the influence of regional processes on local communities. in : Ricklefs R. E., and D. Schluter, editors. Species diversity in ecological communities. University of Chicago Press, Chicago, IL USA. Holway D. A., L. Lach, A. V. Suarez, N. D. Tsutsui, and T.J. Case. 2002. The causes and consequences of ant invasions. A nnual Review of Ecology and Systematics 33:181 233. Holyoak, M. 2000. Habitat subdivision causes changes in food web s tructure. Ecology Letters 3:509 515. Hortal J., P. A. V. Borges, and C. Gaspar. 2006. Evaluating the performance of species richness esti mators: Sensitivity to sample grain size. Journal of Animal Ecology 75:274 287. Hurlbert S H 1971. The nonconcept of species diversity: A critique and alternative parameters. Ecology 52:577 589. Intergovernmental Panel on C limate Change. 2007. Climate Change 2007: Synthesis Report. Cambridge University Press, Cambridge, UK Jose, S., E. J. Jokela, and D. Miller. 2006. The longleaf pine ecosystem: ecology, silviculture, and restoration. Springer, New York, NY USA. Kardol, P., C. E. Campany, L. Souza, R. J. Norby, J. F. Weltzin, and A. T. Classen. 2010. Climate change effects on plant biomass alter dominance patterns and community evenness in an experimental oldfield ecosystem. Global Change Biology 16:2676 2687. Kaspari, M., C. Chang, and J. Weaver. 2010. Salted roads and sodium limitation in a northern forest ant community. Ecological Entomology 35:543 548.
80 Kaspari, M., S. P. Yanoviak, and R. Dudley. 2008. On the biogeography of salt limitation: a study of ant communities. Proceedings of the National Academy of Sciences ( USA ) 105:17848 17851. Kaspari, M., S. P. Yanoviak, R. Dudley, M. Yuan, and N. A. Clay. 2009. Sodium shortage as a constraint on the carbon cycle in an inland tropical forest. Proceedings of the National Academy of Sciences ( USA ) 106 :19405 19409. Kaspari, M., M. Yuan, and L. Alonso. 2003. Spatial grain and the causes of regional diversity gradients in ants. American Naturalist 161:459 477. King, J. R. and W. R. Tschinkel. 2006. Experimental evidence that the introduced fire ant, Sole nopsis invicta does not competitively suppress cooccurring ants in a disturbed habitat. The Journal of Animal Ecology 75:1370 1378. King, J. R. and W. R. Tschinkel. 2008. Experimental evidence that human impacts drive fire ant invasions and ecological change. Proceedings of the National Academy of Sciences ( USA ) 105:20339 203 43. King, J. R., W. R. Tschinkel, and K. G. Ross. 2009. A case study of human exacerbation of the invasive species problem: transport and establishment of polygyne fire ants in Tall ahassee, Florida, USA. Biological Invasions 11:373 377. Komonen, A., R. Penttila, M. Lindgren, and I. Hanski. 2000. Forest fragmentation truncates a food chain based on an oldgrowth forest bracket fungus. Oikos 90:119 126. Korzukhin, M. D., S. D. Porter, L. C. Thompson, and S. Wiley. 2001. Modeling temperaturedependent range limits for the fire ant Solenopsis invicta (Hymenoptera: Formicidae) in the United States. Environmental Entomology 30:645 655. Kruess, A. and T. Tscharntke. 1994. Habitat fragmentation, species loss, and biological control. Science 264:1581 1584. Laurance, W. F., T. E. Lovejoy, H. L. Vasconcelos, E. M. Bruna, R. K. Didham, P. C. Stouffer, C. Gascon, R. O. Bierregaard, S. G. Laurance, and E. Sampaio. 2002. Ecosystem decay of Amazonian forest fragments: A 22year investigation. Conservation Biology 16:605 618. Layman, C. A., J. P. Quattrochi, C. M. Peyer, and J. E. Allgeier. 2007a. Niche width collapse in a resilient top predator following ecosystem fragmentation. Ecology Lett ers 10:937 944. Layman, C. A., D. A. Arrington, C. G. Montana, and D. M. Post. 2007b. Can stable isotope ratios provide for community wide measures of t rophic structure? Ecology 88:42 48.
81 Littell, R. C., G. A. Milliken, W. W. Stroup, R. D. Wolfinger, and O. Schabenberger. 2006. SAS for Mixed Models (SAS Institute Inc, Cary, NC, ed. 2). Lofgren, C. S., W. A. Banks, and B. M. Glancey. 1975. Biology and control of imported fire ants. Annual Review of Entomology 20:130. Lubertazzi D. and W. R. Tschinkel 2003. Ant community change across a ground vegetation gradient in north Floridas longleaf pine flatwoods. Journal of Insect Science 3: art 21. Mac k, R. N., D. Simberloff, W. M. Lonsdale, H. Evans, M. Clout, and F. A. Bazzaz. 2000. Biotic invasions: causes, epidemiology, global consequences, and control. Ecological Applications 10:689 710. Macom T. E. and S. D. Porter. 1996. Comparison of polygyne and monogyne red imported fire ant (Hymenoptera: Formicidae) population densities. Annals of the Entomologi cal Society of America 89:535 543. McCann, K. S. 2000. The diversity stability debate. Nature 405:228 233. McHugh, P. A., A. R. McIntosh, and P. G. Jellyman. 2010. Dual influences of ecosystem size and disturbance on food chain length in streams. Ecology Letters 13:881 890. Minor, E. S., S. M. Tessel, K. A. M. Engelhardt, and T. R. Lookingbill. 2009. The role of landscape connectivity in assembling exotic plant communities: a network analysis. Ecology 90:1802 1809. Moise, E. R. D. and H. A. L. Henry. 2010. Like moths to a street lamp: exaggerated animal densities in plot level global change field experiments. Oikos119:791 795. Moller H. 1996. Lessons for invasion theory from social insects. Biological Conservation, 78:125142. Mooney, K. A. and C. V. Tillberg. 2005. Temporal and spatial variation to ant omnivory in p ine forests. Ecology 86:1225 1235. Morrison L. W. 2002. Long term impacts of an arthropodcommunity invasion by the imported fire ant, Solenopsis invicta. Ecology 83:2337 2345. Morrison, L. W., M. D. Korzukhin, and S. D. Porter. 2005. Predicted range expansion of the invasive fire ant, Solenopsis invicta, in the Eastern United States based on the VEMAP global warming scenario. Diversity and Distributions 11:199 204. National Atmo spheric Deposition Program 2011. Sodium ion wet deposition. Available at: http://nadp.sws.uiuc.edu/maplib/pdf/2011/Na_dep_11.pdf. Accessed January 22, 2013.
82 Newsome, S. D., C. Martnez del Rio, S. Bearhop, and D. L. Phillips. 2007. A niche for isotopic ecology. Frontiers in Ec ology and the Environment 5:429 436. Noss, R. F. 1989. Longleaf pine and wiregrass: Keystone components of an endangered ecosyst em. Natural Areas Journal 9:234 235. ODonnell, S. et al. 2010. Leaf cutter ants ( Atta cephalotes ) harvest baits offering sodium chloride rewards. Insectes Sociaux 57:205 208. Oksanen J., et al. 2010. Vegan: Community ecology package (R package version 2.02). Orrock J. L., G. R. Curler, B. J. Danielson and D. R. Coyle. 2011. Largescale experimental lands capes reveal distinctive effects of patch shape and connectivity on arthropod communities. Landsc ape Ecology 26:1361 1372. Orrock, J. L., and E. I. Damschen. 2005. Corridors cause differential seed predation. Ecological Applications 15:793 798. Palmer, T. M. 2003. Spatial habitat heterogeneity influences competition and coexistence in an African ac acia ant guild. Ecology 84:2843 2855. Parmesan, C. 2006. Ecological and Evolutionary Responses to Recent Climate Change. Annual Review of Ecology, Evolution, and Systematics 37:637 669. Pelini, S. L., F. P. Bowles, A. M. Ellison, N. J. Gotelli, N. J. Sanders, and R. R. Dunn. 2011a Heating up the forest: opentop chamber warming manipulation of arthropod communities at Harvard and Duke Forests. Methods in Ecology and Evolution 2:534 540. Pelini, S. L., M. Boudreau, N. McCoy, A. M. Ellison, N. J. Gotelli, N. J. Sanders, and R. R. Dunn. 2011b Effects of short term warming on low and high latitude forest ant communities. Ecosphere 2 art62. Pimm, S. L. 2002. Food webs. University of Chicago Press, Chicago, Illinois, USA. Porter, S. D. 1988. Impact of temperature on colony growth and development rates of the ant, Solenopsis invicta. Journal of Insect Physiology 34:1127 1133. Porter, S. D. 1989. Effects of diet on the growth of laboratory fire ant colonies (Hymenoptera: Formicidae). Journal of the Kansas Entomological Society 62:288 291. Porter, S and D. Savignano. 1990. Invasion of polygyne fire ants decimates native ants and disrupts arthropod community. Ecology 71:2095 2106. Porter, S. D., and W. R. Tschinkel. 1987. Foraging in Solenopsis invicta (Hymenoptera: Formicidae): effects of weather and season. Environmental Entomology 16: 802 808.
83 Post, D. M. 2002. Using stable isotopes to estimate trophic position: Models, methods, and assumptions. Ecology 83:703 718. Post, D. M., M. L. Pace, and N. G. Hairston. 2000. Ecosystem size determines foodchain length in lakes. Nature 405:1047 1049. Proches, S., J. Wilson, R. Veldtman, J. Kalwij, and D. Richardson. 2005. Landscape corridors: Po ssible dangers? Science 310:779 783. R Development Core Team. 2009. R: a language and environment for statistical computing. Vienna, Austria. R Development Core Team. 2011. R: A language and environment for statistical computing Vienna, Austria. Resasco J., D. J. Levey, and E. I. Damschen. 2012. Habitat corridors alter trophic position of fire ants. Ecosphere, 3, ar t. 11. Ries, L., R. J. Fletcher, J. Battin, and T. D. Sisk. 2004. Ecological responses to habitat edges: Mechanisms, models, and variability explained. Annual Review of Ecology Evolution and Systematics 35:491522. Robinson, W. D. 1999. Long Term Changes i n the Avifauna of Barro Colorado Island, Panama, a Tropical Forest Isolate. Conservation Biology 13:85 97. Ross, K. G., and L. Keller. 1998. Genetic control of social organization in an ant. Proceedings of the National Academy of Sciences ( USA ) 95:14232 1 423 7. Ruano, F., A. Tinaut, and J. J. Soler. 2000. High surface temperatures select for individual foraging in ants. Behavioral Ecology 11:396 404. Sagarin, R. D., J. P. Barry, S. E. Gilman, and C. H. Baxter. 1999 Climate related change in an intertidal community over short and long time scales. Ecological Monographs 69:465 490. Sanders, N. J., J. P. Lessard, M. C. Fitzpatrick, and R. R. Dunn. 2007. T emperature, but not productivity or geometry, predicts elevational diversity gradients in ants across spatial grains. Global Ecology and Biogeography 16:640 649. Schmidt, S. N., J. D. Olden, C. T. Solomon, and M. J. Vander Zanden. 2007. Quantitative approaches to the analysis of stable isotope food web data. Ecology 88:2793 2802. Schoener, T. W. 1989. Food webs from the small to the large. Ecol ogy 70:1559 1589. Shoemaker D., and M. S. Ascunce. 2010. A new method for distinguishing colony social forms of th e fire ant, Solenopsis invicta. Journal of Insect Science 10 :art 10.
84 Siemann, E., D. Tilman, J. Haarstad, and M. Ritchie. 1998. Experimental tests of the dependence of arthropod diversity on plant diversi ty. American Naturalist 152:738 750. Suarez, A. V., D. A. Holway, and T. J. Case 2001. Patterns of spread in biological invasions dominated by long distance jump dispersal: Insights from Argentine ants. Proceedings of the National Academy of Sciences (U S A ) 98 :1095 1100. Simberloff, D. and J. Cox. 1987. Consequences and costs of conservation corridors. Conservation Biology 1:63 71. Simberloff, D., J. A. Farr, J. Cox, and D. W. Mehlman. 1992. Movement Corridors: Conservation Bargains or Poor Investments? Conservation Biology 6:493 504. S mith, V. G. 1928. Animal communities of a deciduous f orest succession. Ecology 9:479 500. Smith, J. R., P. Fong, and R. F. Ambrose. 2006. Dramatic declines in mussel bed community diversity: response to climate change? Ecology 87:1153 1161. Stiles J. H. and R. H. Jones 1998. Distribution of the red imported fire ant, Solenopsis invicta in road and powerline habitats. Landscape Ecology 13: 335 346. Stamp N. E. and G. D. Harmon. 1991. Effect of potassium and sodium on fecundity and survivorship of Japanese beetles. Oikos, 62:299 305. Strayer D. L., V. T. Eviner, J. M. Jeschke, and M. L. Pace. 2006. Understanding the long term effects of species invasions. Trends in Ecology and Evolution 21:645 651. Stuble, K. L., S. L. Pelini, S. E. Diamond D. A. Fowler R. R. Dunn and N. J. Sanders 2013. Foraging by forest ants under experimental climatic warming: a test at two sites Ecology and Evolution 3:482 491 Sullivan, L. L., B. L. Johnson, L. A Brudvig, and N. M. Haddad. 2011. Can dispersal mode predict corridor effects on plant parasites? Ecology 92:1559 1564. Sutcliffe, O. L. and C. D. Thomas. 1996. Open corridors appear to facilitate dispersal by ringlet butterflies ( Aphantopus hyperantus ) between woodland clearings. Conservation Biology 10:1359 1365. Takim oto, G., D. A. Spiller, and D. M. Post. 2008. Ecosystem size, but not disturbance, determines foodchain length on islands of the Bahamas. Ecology 89:3001 3007. Tewksbury, J. J., D. J. Levey, N. M. Haddad, S. Sargent, J. L. Orrock, A. Weldon, B. J. Daniels on, J. Brinkerhoff, E. I. Damschen, and P. Townsend. 2002. Corridors affect plants, animals, and their interactions in fragmented landscapes. Proceedings of the National Academy of Sciences (USA) 99:12923 12926.
85 Tillberg, C. V., D. A. Holway, E. G. LeBrun, and A. V. Suarez. 2007. Trophic ecology of invasive Argentine ants in their native and introduced ranges. Proceedings of the National Academy of Sciences (USA) 104:20856 20861. Tillberg, C. V., D. P. McCarthy, A. G. Dolezal, and A. V. Suarez. 2006. Measur ing the trophic ecology of ants using stable i sotopes. Insectes Sociaux 53:65 69. Tschinkel, W R 1998. The reproductive biology of fire ant societies. BioScience 48:593 605. Tschinkel, W. R. 2006. The fire ants. The Belknap Press of Harvard University Press, Cambridge, Massachusetts, USA. Valles S. M. and S. D. Porter. 2003. Identification of polygyne and monogyne fire ant colonies ( Solenopsis invicta ) by multiplex PCR of Gp 9 alleles. Insectes Soc iaux 50:199 2 00. Valles, S. M. and S. D. Porter. 2013. Procedures to mitigate the impact of Solenopsis invicta virus 3 in fire ant rearing facilities. Florida Entomologist 96: 252 254. Van Lear, D. H., W. D. Carroll, P. R. Kapeluck, and R. Johnson. 2005. History and restoration of the longleaf pinegrassland ecosystem: Implications for species at risk. Forest Ecology and Management 211:150 165. Vander Zanden, M. J., B. J. Shuter, N. Lester, and J. B. Rasmussen. 1999. Patterns of food chain length in lakes: A stable is otope study. American Naturalist 154:406 4 41 Yang C.C., et al. 2012. Propagule pressure and colony social organization are associated with the successful invasion and rapid range expansion of fire ants in China. Molecular Ecology 21:817 833. Vail, K. M., D. F. Williams, and D. H. Oi 1999. Ant bait attractive to multiple species of ants. Awarded August 17, 1999. United States Patent #5,939,061. Van Lear, D. H., W. D. Carroll, P. R. Kapeluck, and R. Johnson. 2005. History and restoration of the longleaf pinegrassland ecosystem: Implications for species at risk. Forest Ecology and Management 211:150 165. Van Pelt, A. F. and J. B. Gentry. 1985. The ants (Hymenoptera: Formicidae) of the Savannah River Plant, South Carolina. SRO NERP 14. Vargo E. L., and S. D. Porter 1989. Colony reproduction by budding in the polygyne form of Solenopsis invicta (Hymenoptera: Formicidae). Annals of the Entomological Society of America 82:307 313. Vitousek, P. M., H. A. Mooney, J. Lubchenco, and J. M. Melillo. 1 997. Human d omination of earths e cosystems. Science 277:494 499.
86 Vogt J. T., M. L. Allen, B. Wallet, D. Boykin, and W. A. Smith. 2009. Distribution patterns of imported fire ants (Hymenoptera: Formicidae) on a sheep and goat farm in Oklahoma. Environment al Entomology 38:551 560. Walker, M. D. et al. 2006. Plant community responses to experimental warming across the tundra biome. Proceedings of the National Academy of Sciences (USA) 103:1342 1346. Weldon, A. J. 2006. How corridors reduce indigo bunting nest success. Conservation Biology 20:1300 1305. Whitford, W. G., J. Van Zee, M. S. Nash, W. E. Smith, and J. E. Herrick. 1999. Ants as indicators of exposure to environmental stressors in North Amer ican desert grasslands. Environmental Monitoring and Assessment 54:143 171. Wilcove, D. S., D. Rothstein, J. Dubow, A. Phillips, a nd E. Losos. 1998. Quantifying threats to imperiled s pecies in the United States. BioScience 48:607 615. Woodcock, P., et al. 2012. Assessing trophic position from nitrogen isotope ratios: effective calibration against spatially varying baseli nes. Naturwissenschaften 99:275 2 83.
87 BIOGRAPHICAL SKETCH Jul ian Resasco received his BS in z oology at the University of Oklahom a in 2006. After graduation he worked as a field technician at the Savannah River Site, SC in the Corridor Project. He received an NSF Graduate Research Fellowship and did his PhD in the Department of Biology at t he University of Florida He received his PhD in summer of 2013 and began his postdoctoral work as an NSF Postdoctoral Research Fellow in Biology based out of the University of Colorado Boulder.