1Biology Department, College of Charleston, Charleston, SC 29424. 2Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT 06269. *Corresponding author - scholtensb@cofc.edu. Lepidoptera of Great Smoky Mountains National Park: Methods and Results of the Inventory Brian G. Scholtens1,* and David L. Wagner2 Abstract - As a part of an All Taxa Biodiversity Inventory, we documented 1843 Lepidoptera species from Great Smoky Mountains National Park. This total is based on previous survey efforts plus recent intensive sampling using bioblitzes. Various statistical estimators put the total number of species in the park between 1950 and 2340. The difference between actual and estimated numbers is mainly due to under-sampled Microlepidoptera groups. As a part of recent efforts, the mitochondrial COI gene has been sequenced from more than 940 species from the Park. DNA barcoding has already led to taxonomic insights in several groups, and is believed to be at least 95% accurate for identifications. Our samples include more than 20 undescribed species in the Park, including a park-endemic geometrid moth. Because of threats to their habitats, high-elevation species make up the largest group of species of special concern. Introduction Lepidopterists have been actively involved in the All Taxa Biodiversity Inventory (ATBI) of Great Smoky Mountains National Park (GSMNP) from the inception of the inventory, with Lepidoptera designated as one of the original Taxonomic Working Groups (TWiGs). An organizational meeting of interested scientists was held in December of 1997 during the first Discover Life in America (DLIA) annual meeting, and survey work began during 1999. Lepidopterists felt that they could have an immediate impact because of the advanced state of the order’s taxonomy, the limited knowledge of the Lepidoptera in the Park, and a core of lepidopterists in close proximity willing to volunteer their time, resources, and expertise. Conducting a survey of a diverse insect taxon such as the Lepidoptera poses significant challenges, and we understood from the beginning that our effort in Great Smoky Mountains National Park would be a long-term effort. Like other insects, butterflies and moths are often habitat specific and/or tied to a particular host plant. Many have restricted flight periods as adults. Some are so small or specialized that they are not likely to be encountered using standard light-trapping or baiting techniques. All these characteristics make a comprehensive survey a difficult task. Powell (1995) found no complete North American Lepidoptera surveys at the 193 The Great Smoky Mountains National Park All Taxa Biodiversity Inventory: A Search for Species in Our Own Backyard 2007 Southeastern Naturalist Special Issue 1:193–206 194 Southeastern Naturalist Special Issue 1 regional, state, provincial, or even local level. He regarded the most complete state surveys to include those for New York (Forbes 1923–1960) and Maine (Brower 1974–1986). Adams (2002) conducted the most intensive survey of a locality near GSMNP, documenting 758 species from the area around Highlands, NC. These are invariably multi-year surveys, using many collection and observation techniques. We assumed that our work would face similar hurdles. The Lepidoptera TWiG’s goals have been to compile a comprehensive species list of Lepidoptera for the Park, better understand the distributions of Lepidoptera within the Park, bring attention to species of potential concern within the Park, and aid the Park in making management decisions. The survey was expected to have the side benefit of promoting systematic research by providing specimens, many of which, especially among the Microlepidoptera, were poorly represented in collections. We also anticipated it would reveal new life-history data, e.g., host plant, habitat, and parasitoid associations, and as a consequence, promote behavioral, ecological, and other studies involving Lepidoptera. Other benefits of the efforts of the Lepidoptera TWiG included: actively training and enlisting the help of volunteers and students; interacting in numerous ways with students at the K–12, undergraduate, graduate, and post-doctoral levels; and engaging in numerous public outreach programs. We will show that these and other benefits have been realized, and that similar results likely will be obtained from other surveys of this kind. History of the Lepidoptera Inventory in Great Smoky Mountains National Park Great Smoky Mountains National Park was established in 1934 to protect a portion of the spectacular southern Appalachian Mountains, their unique fl ora and fauna, and the communities in which they occur. One of the first Park naturalists, Art Stupka, was interested in recording the diversity of organisms in the Park. During his long tenure, he collected many Lepidoptera, largely butterfl ies (50 spp.), but also many moths (36 spp.). Several other collectors visited the Park during the 1930s and 1940s and deposited specimens in the Park collection (or had their collections deposited there at a later date). The most significant were M. Sullivan (43 spp. vouchered from NC), F.C. Evans (28 spp.), and Earl R. Cady (21 spp.). During the 1950s, 60s, and 70s, the most significant contributions were made by K. Hobson (69 spp.), K.D. Snyder (49 spp.), and P.W. Pfitzer (37 spp.). During the 1980s, activity accelerated (Fig. 1). Park employees Keith Langdon (97 spp.) and Don DeFoe (43 spp.) both deposited significant numbers of Lepidoptera in the Park collection, and Richard Brown conducted a survey of moths from the Park (Brown 1986). At the end of the 1980s, the first focused survey was initiated by Paris Lambdin et al. (1991). Sampling occurred at the same localities throughout the two-year survey; nearly all specimens vouchered and identified 2007 B.G. Scholtens and D.L. Wagner 195 were Macrolepidoptera. Many of these specimens are deposited in the Park collection, but another set is at the University of Tennessee. Lambdin et al. (1991) listed just fewer than 800 species, 349 of which previously had not been recorded in the park. DeFoe and Langdon continued collecting through the 1990s. Langdon also conducted a series of butterfl y surveys in Cades Cove using Pollard walk techniques (Pollard and Yates 1994), vouchering specimens when necessary. Methods and Materials Current survey efforts From its inception, ATBI sampling took two forms: 1) structured, plotbased sampling with Malaise, pit fall, and fl ight-intercept traps, which was carried out over a period of 2 years; and 2) opportunistic sampling based on the interests of individual researchers. Individuals conducted initial sampling of Lepidoptera during the ATBI. Beginning in May 1999, Brian Scholtens collected moths near Tremont, in Cades Cove, and from various sites in the vicinity of Cosby. In 1999, Eric Metzler received an early grant from DLIA to sample late-season Lepidoptera (Metzler 1999). Subsequently, many other lepidopterists became involved in the survey. Supported by DLIA grants, John Brown and Don Davis carried out extensive blacklight bucket-trap sampling at several localities in 2001 and 2002, and Michael Pogue, using DLIA volunteers and llama teams, conducted the first extensive light-trap sampling on Figure 1. Accumulation of Lepidoptera species records in Great Smoky Mountains National Park by year, from 1926 to 2006. 196 Southeastern Naturalist Special Issue 1 high-elevation balds. J. Bolling Sullivan, under contract to the National Park Service (NPS), intensively sampled the Oconaluftee area as a part of an environmental impact study for a property that subsequently was transferred to the Cherokee tribe (Sullivan and Deutschmann 2002). In addition, Paul Super and his students, while working at the Great Smoky Mountains Institute at Tremont, initiated a moth survey in the first year of the ATBI. After Super left Tremont, Michelle Prysby and visiting students continued the inventory. Super later extended his survey to the Appalachian Highlands Learning Center at Purchase Knob in North Carolina. Both of these latter projects have made significant contributions to the Lepidoptera survey, particularly of early and late-season species. More recently, in 2004, J. Bolling Sullivan again was contracted by the NPS to survey multiple locations along the north shore of Fontana Lake, mostly at remote sites requiring access by boat (Sullivan 2004). Lepidopterists have not taken advantage of the structured, plot-based sampling because of the difficulty in handling specimens in alcohol from Malaise-trap samples and the relative inefficiency of this passive collecting technique for Lepidoptera. Nevertheless, a handful of significant records have come from these samples. Two in particular are the first North Carolina and Park record of Euchloe olympia (Edwards) (Olympia Marble: Pieridae) and the first Park records of the primitive, mandibulate species Epimartyria auricrinella Walsingham (Micropterigidae). Our first organized effort was a bioblitz in July of 2000, spearheaded by David Wagner, who had been involved in similar events in Connecticut designed to bring considerable taxonomic expertise together for intensive 1- to 2-day sampling efforts. For the first Lepidoptera bioblitz, we drew together 15–20 scientists from across the continent and many park staff and volunteers to inventory a large number of sites across the Park. During the bioblitz, we made an effort to sample all major habitats and to collect across the Park’s elevational range, using up to 40 traps per night. We estimate that we sampled and sorted in excess of 20,000 moths over a 24-hour period. Downsides to this sampling technique include time spent coordinating many volunteers and the loss of many small Microlepidoptera because they cannot be sorted and prepared quickly enough. To efficiently process samples, gather vouchers, and construct a list, we organized into subTWiGs. The subTWiGs were: 1) primitive Lepidoptera, 2) Gelechioidea, 3) Tortricoidea, 4) Pyraloidea and Pterophoroidea, 5) Hesperioidea and Papilionoidea, 6) Zygaenoidea, 7) Geometroidea and Drepanoidea, 8) Bombycoidea and Sphingoidea, and 9) Noctuoidea. Each subTWiG was assigned one or more leaders (depending on size of the group and interest). Leaders were responsible for gathering vouchers from all samples to cover the diversity of their group(s) and keeping a comprehensive list of species (or morphospecies). The bioblitz began at 3:00 pm and ended the following day, with lists completed by 3:00 pm. All sampling, sorting, and preliminary listing were completed in a 24-hour period. 2007 B.G. Scholtens and D.L. Wagner 197 Lists were put together using pre-made checklists of probable species and tabulated using a Filemaker Pro database to construct the final inventory list. Individual specimens were retained by specialists for later preparation, dissection, and/or confirmation of identifications. As preparation and identifications were completed, the data were sent to Brian Scholtens and entered in a Filemaker Pro database. The basic organization of the first bioblitz was retained in subsequent efforts, but we modified our efforts to voucher more completely all species encountered, and added a second day for specimen processing and data capture. Beginning with the second bioblitz, we made special efforts to capture as many geo-referenced occurrence records as possible during the 2-day run of the bioblitz. Follow-up bioblitzes occurred in May 2001, June 2002, July 2004, May 2005, and August 2006. Starting in 2004, we augmented our bioblitz protocol with efforts to gather a mitochondrial DNA sample from each species. These samples were processed in Paul Hebert’s lab at the University of Guelph to generate sequence data for approximately 650 base pairs of the COI gene of each species (Hebert et al. 2003, Janzen et al. 2005). During the 2004 and 2005 bioblitzes, we also provided specimens to the Ambrose Monnell collection of tissue samples at the American Museum of Natural History for cryopreservation. These specimens are stored in liquid nitrogen to provide a permanent source of tissue samples for future research efforts. Results Completeness of species list Work from past decades plus our current work on the ATBI has brought the list of species known from the Park to 1843, more than doubling the number known from the Park since the start of the ATBI. The initial bioblitz in 2000 added over 350 species to the Park list (again, in just 24 hours). Another 150 were added in 2002, with diminishing numbers in subsequent efforts (Fig. 1). Recent bioblitzes and other sampling efforts within the Park have added few new taxa of larger moths and butterfl ies, suggesting that our sampling is 95–100% complete for most Macrolepidoptera groups (Bombycoidea, Sphingoidea, Geometroidea, Noctuoidea, Papilionoidea, Hesperioidea; see also below). Several Microlepidoptera groups (e.g., Zygaenoidea, Pyraloidea, and Tortricoidea) also are well sampled because of the participation of experts, the comparatively large size of the moths, and the advanced state of knowledge of their taxonomy. Other Microlepidoptera groups are still inadequately inventoried, either because experts have not been involved to the same degree, or because special efforts are necessary to detect the species (e.g. rearing, pheromone trapping, etc.). The number of species known from each family of Lepidoptera is presented in Table 1. We used several methods to estimate the number of species of Lepidoptera that likely occurs in the Park. At the first bioblitz, the participating scientists were each asked to estimate how many species of their 198 Southeastern Naturalist Special Issue 1 taxon would eventually be found within the Park. Based on these bestguess extrapolations the total for the order was estimated to be between 2500 and 3000. With much more extensive collecting now completed we can estimate the total more accurately. It is difficult to apply most statistical estimators because we have not routinely recorded all species in all samples, and samples (= effort) have not been equivalent (see Brown and Bash 1997, Powell 1995). Because sampling efforts have been relatively constant, or at least relatively comprehensive, from year to year during the Table 1. Number of species of Lepidoptera known from Great Smoky Mountains National Park (taxonomy follows Kristensen (1999)). Family # of species Family # of species Family # of species Micropterigidae 1 Cosmopterigidae 10 Megalopygidae 3 Eriocraniidae 1 Xyloryctidae 1 Limacodidae 21 Hepialidae 4 Gelechiidae 52 Crambidae 115 Nepticulidae 12 Schreckensteiniidae 1 Pyralidae 106 Opostegidae 1 Epermeniidae 1 Thyrididae 3 Tischeriidae 10 Lyonetiidae 1 Pterophoridae 17 Incurvariidae 1 Glyphipterigidae 3 Drepanidae 7 Prodoxidae 3 Yponomeutidae 8 Geometridae 225 Adelidae 2 Plutellidae 1 Uraniidae 2 Heliozelidae 6 Acrolepiidae 2 Mimallonidae 2 Tineidae 33 Sesiidae 17 Bombycidae 2 Acrolophidae 9 Cossidae 3 Saturniidae 13 Psychidae 3 Tortricidae 218 Sphingidae 30 Bucculatricidae 5 Choreutidae 4 Lasiocampidae 7 Gracillariidae 61 Hesperiidae 42 Notodontidae 40 Oecophoridae 12 Papilionidae 6 Arctiidae 49 Amphisbatidae 4 Pieridae 12 Lymantriidae 10 Elachistidae 19 Lycaenidae 19 Pantheidae 5 Glyphidoceridae 3 Nymphalidae 32 Nolidae 16 Coleophoridae 13 Zygaenidae 3 Noctuidae 528 Autostichidae 2 Epipyropidae 1 Total 1843 Table 2. Estimated species numbers for Lepidoptera groups based on Chao-2 estimator (observed number is from 1999–2004 sampling; in parentheses are known totals for all years of sampling) Taxon Observed species # Expected species # % complete Macrolepidoptera Saturniidae 12 (13) 13 92 (100) Sphingidae 25 (30) 30 83 (100) Noctuidae/Pantheidae/Nolidae 488 (549) 542 90 (101) Arctiidae 41 (49) 43 95 (114) Notodontidae 38 (40) 40 95 (100) Geometridae 211 (225) 255 83 (88) Papilionoidea/Hesperioidea 61 (111) 64 95 (173) Microlepidoptera Zygaenoidea 25 (28) 27 93 (104) Tortricidae 200 (218) 241 83 (90) Pyraloidea 217 (221) 246 88 (90) 2007 B.G. Scholtens and D.L. Wagner 199 current survey, we can use each year as a sample for the Chao-2 estimator (Table 2). For the 6 sample years of 1999–2004, the Chao-2 estimator predicts that the inventories for most groups of Macrolepidoptera are 83–95% complete over the course of those 6 years. If known species are added from earlier sampling efforts, the number of species in most families exceeds the Chao-2 estimated richness, suggesting that we are approaching 100% completion for these groups (Table 2). We are undoubtedly missing a small number of resident species, especially from large families. As importantly, because of faunal changes over the history of the Park, our totals may be an overestimate of extant diversity. By assuming that the ratio of Macrolepidoptera/Microlepidoptera in Great Smoky Mountains National Park is similar to the ratio for all of North America (i.e., approximately 1:1; Hodges et al. 1983) it is possible to estimate of the total number of Microlepidoptera that should occur in the Park. Table 3 gives extrapolations for Microlepidoptera richness estimates for the Park, based on assumptions of 90, 95, and 100% completeness. Adding predicted richness totals for both Macrolepidoptera and Microlepidoptera, our estimates for total GSMNP richness range from 2106 (given 100% of the Macrolepidoptera have been recorded) to 2339 (assuming that 90% of the Macrolepidoptera have been recorded). These extrapolations suggest that our Microlepidoptera survey is 68–75% complete. We used a similar procedure to produce estimates based on a list of the Lepidoptera of the state of Connecticut (D. Wagner, unpubl. data; Table 4). The Connecticut list is believed to be the most taxonomically comprehensive of any state, and especially so for more than 30 families of Microlepidoptera, with only the Gelechiidae known with appreciable uncertainty. Assuming Table 3. Comparison of North America and Great Smokies National Park (GSMNP) Lepidoptera faunas (% completion is based on sampling of Macrolepidoptera used to scale Microlepidoptera; North American species richness from Hodges et al. 1983). 90% 95% 100% North complete complete complete America GSMNP GSMNP/NA estimate estimate estimate Macrolepidoptera 5814 1047 18.0% 1163 1102 1047 Microlepidoptera 5878 796 13.5% 1176 1114 1059 Total 11692 1843 15.8% 2339 2216 2106 Table 4. Comparison of Connecticut and Great Smokies National Park (GSMNP) Lepidoptera faunas (% completion is based on sampling of Macrolepidoptera used to scale Microlepidoptera; Connectitcut species richness from D. Wagner, unpubl. data). 90% 95% 100% complete complete complete Connecticut GSMNP GSMNP/CT estimate estimate estimate Macrolepidoptera 1187 1047 88.2% 1163 1102 1047 Microlepidoptera 1022 796 77.9% 1001 949 901 Total 2209 1843 83.4% 2164 2051 1948 200 Southeastern Naturalist Special Issue 1 our Macrolepidoptera sampling is 90, 95, or 100% complete, we estimate the total number of Lepidoptera species to be between 1948 and 2164 and that our GSMNP Microlepidoptera list is 80–88% complete. While the fauna would not be expected to exactly mirror any particular geographical region, we use these ratios because Connecticut has similar plant communities and no other comparably complete surveys are available. All of the estimates place the GSMNP species total between about 1950 and 2350 species, with the lower estimates apparently more likely based on Chao-2 estimators of the completeness of our Macrolepidoptera sampling. Significant species The Lepidoptera fauna of Great Smoky Mountains National Park is dominated by species typical of the eastern deciduous forest, especially the cove forests of the Park’s middle and lower elevations. Fewer species are associated with high-elevation spruce/fir forests, pine/oak woodlands, wetlands, and open areas. Missing or poorly represented species are associated with emergent aquatic vegetation (especially nymphuline pyralids), sandy soils, highly disturbed habitats, fl ood plain or bottomland forests, expansive barrens, acidic wetlands, heathlands, and large grasslands, all habitats that are poorly represented in the Park. The most significant species of conservation interest are endemics and disjuncts. Several endemics are limited to the Park or the middle and southern Appalachian region. At least two are known from the cove forests, and both are undescribed. Ligdia sp. (Geometridae) was discovered on one of our initial collecting trips. It is known from only two locations in the Park. The host plant is Euonymus obovatus Nuttall, an Appalachian endemic characteristic of mid-elevation boulder fields (D. Wagner, unpubl. data). This species is the first record of its genus and tribe (Abraxini) in the New World. Also known from lower elevation cove forests is an undescribed pyraustine (Crambidae). This species has a wider distribution in the middle and southern Appalachians; its life history is unknown. In addition, Braun (1935) described Phyllonorycter rhododendrella (Braun) (Gracillariidae) from reared material collected in the Park. Nancy Lowe recently rediscovered this species in the Park at its type locality. It is otherwise known only from another locality a little north of Atlanta. There are at least three endemics at high elevations, all confined to the middle to southern Appalachians. Two are apparently associated with Gazoryctra sciophanes (Ferguson) (spruce/fir forests), (Hepialidae) and Anaplectoides brunneomedia McDunnough (brown-lined dart; Noctuidae). The third is an undescribed Agriphila (Crambidae) found in open habitats such as balds and roadsides. These species seem especially vulnerable as spruce/fir habitats are gradually transformed by insect invaders (e.g., balsam woolly adelgid) and global warming. Saving these species may require active management and/or other conservation efforts in the future. Many disjuncts, Polix coloradella (Walsingham) (Oecophoridae), Eulia ministrana (Linnaeus), Apotomops wellingtonana (Kearfott) (both 2007 B.G. Scholtens and D.L. Wagner 201 Tortricidae), Platarctia parthenos (Harris) (St. Lawrence tiger moth; Arctiidae), Cucullia florea Gn., Diachrysia aeroides (Grote) (dark-spotted looper moth), Diachrysia balluca Geyer, Lithophane georgii Grote, Polychrysia morigera (Henry Edwards), Xestia perquiritata (Morrison), Pachypolia atricornis Grote, Platypolia anceps (Stephens) (all Noctuidae), Polygonia faunus (W.H. Edwards) (green comma), and Phyciodes batesii (Reakirt) (tawny crescent) (both Nymphalidae) exist at high elevations in the Park, apparently separated from the closest populations in the central and northern Appalachians after the last glacial period. In addition to the two undescribed species mentioned above, approximately 25–30 other undescribed species are known from GSMNP. Many of these are in families of relatively poorly known Microlepidoptera, such as the Tineidae and Gelechiidae, but some are Geometridae, Noctuidae, and Arctiidae. Many families of Microlepidoptera require additional sampling, especially those in the Gelechioidea. There are few specialists in this diverse group, many gelechioids are specialized in habit, and identifications can be difficult. Other groups, such as Nepticulidae require rearing to survey efficiently, and additional pheromone trapping will be necessary to complete the inventory of Sesiidae. DNA barcoding To date, 4156 specimens, representing 940 species of Lepidoptera from GSMNP, have been DNA barcoded. While this number includes members of most families, the emphasis up to this point has been on larger, easily (reliably) identified taxa. More than 90% of the species for which we possess multiple barcodes show little (less than 0.5%) variation in their COI sequence. A handful of taxa that are treated as distinct species cannot be distinguished using DNA barcodes. Examples include some species in the genera Catocala (Noctuidae), Macaria (Geometridae), Probole (Geometridae), and Acrobasis (Pyralidae). We also noted a number of instances where “species” had considerable variation in their COI barcodes, i.e., with up to 5% sequence variation. In large DNA barcoding surveys of moths in Canada and Costa Rica, the average divergence rate among topologically proximate (i.e., congeneric and contribal taxa) is on the order of 1–5% divergence (Hebert et al. 2004, Janzen et al. 2005). Variation in COI sequences greater than 2% in Lepidoptera indicates the existence of more than one evolutionary unit. We use this approximate figure as a “threshold value” to flag where more taxonomic study, DNA barcoding, and other types of studies might be needed to more accurately delimit the biological species. GSMNP samples include several species with deep splits that are currently recognized as single entities. In the Noctuidae, Pangrapta decoralis Hübner (decorated owlet), and in the Crambidae, Fissicrambus mutabilis (Clemens), both contain sequence variation considerably greater than that between well-recognized species. In another case, barcoding has caused a researcher to rethink the previous lumping of two taxa (Torticidia flexuosa (Grote) and T. pallida 202 Southeastern Naturalist Special Issue 1 (Herrich-Schäffer): Limacodidae) that can be distinguished using DNA barcodes (M.E. Epstein, California Department of Food and Agriculture, Sacramento, CA, pers. comm.). Discussion Bioblitzes are an efficient method to survey Lepidoptera. Each bioblitz was run on a budget of about $5000 (although scientists, park staff, and volunteers donated considerable amounts of time and effort). We estimate the real cost (value) to be $25,000 if all taxonomists were paid for their services, travel costs were fully funded, and other expenditures were counted. In GSMNP, these efforts were spread over several years, largely because of logistical considerations. Our two largest bioblitzes were tied closely to national meetings so that taxonomists could incorporate their visit to the Park into existing travel plans. Because GSMNP organized the first All Taxa Biodiversity Inventory, the Park was fortunate to receive a great deal of attention from the community of systematic entomologists. Location of the Park also played a role; Great Smoky Mountains National Park is within driving distance for a great many amateur and professional lepidopterists. As additional inventories are initiated at more remote parks, taxonomic expertise likely will be stretched as researchers choose how and where to invest their time. In particular, there may not be enough microlepidopterists to support multiple ATBIs at the same time. The problem will be exacerbated for other orders where there are still fewer practicing taxonomists and amateurs. ATBIs elsewhere may want to emphasize different groups of organisms at different times in order to efficiently use the available (and especially nearby) pools of taxonomic expertise. Even with these caveats, large-scale surveys provide a great return on a relatively small investment. In addition to the distribution, natural history, and conservation information gained, our bioblitzes have provided valuable specimens and information for systematists across the continent. Bioblitzes provide individual researchers with the opportunity to examine many trap catches from a variety of elevations and habitats, and yield a diversity and quantity of specimens that otherwise would be prohibitively time consuming to obtain as an individual researcher. In addition, bioblitzes promote interaction among researchers and between researchers and graduate students, and contribute to the training of the next generation of systematists. We discovered, in some cases, that our bioblitzes were the only time that world experts in different groups of Lepidoptera had ever been in the field together. Bioblitzes provided a unique opportunity for graduate and undergraduate students to study with these experts. These interactions made the bioblitzes important professional experiences for all participants, regardless of the level of their training. One cannot overstate the value bioblitzes provide for any biotic inventory—they yield the “biggest bang for the buck” in regard to the number of localities sampled, specimens generated, species 2007 B.G. Scholtens and D.L. Wagner 203 bar-coded, and geo-referenced occurrence records entered into the Park’s database per unit person or dollar investment. The impact of the Lepidoptera survey has been immediate and important for the Park. We have recognized several species of conservation concern and identified areas within the Park that harbor these endemic, disjunct, or unusual species. The species associated with cove forest habitats are apparently secure, but threats to several trees species and the possibility of competition from invasives warrant special attention. Fagus grandifolia Ehrhart (American beech), Tsuga canadensis (Linnaeus) Carr (eastern hemlock), Quercus spp. (oaks), and Cornus fl orida Linnaeus (fl owering dogwood) are all suffering from or threatened by diseases or insect pests. Many Lepidoptera use these species as hosts, and their loss would have a signifi- cant impact on diversity in the cove-forest habitat. The effect on moths of the earlier loss of Castanea dentata (Marshall) Borkhausen (American chestnut) went unmonitored. Our current survey provides a baseline to assess future changes in the composition of the Lepidoptera fauna. A unique fauna is associated with dry pine/oak forests that are concentrated on the Foothills Parkway at the east and west ends of the Park. A diverse assemblage, including many Noctuidae, Pyralidae (especially Phycitinae), Tortricidae, and Gelechiidae, are found in these habitats and rarely, if ever, in other parts of the Park. Although widespread in other areas, these habitats are restricted in the Park. Potential threats to this habitat include the disease sudden oak death and the invasion of the area by Lymantria dispar (Linnaeus): Lymantriidae (gypsy moth). Probably the most endangered habitats in the Park are the high-elevation spruce/fir forests and balds. These areas act as sky islands in the southern Appalachians and harbor some of the most unique species in the Park (see above). This habitat has been and continues to be assaulted by invasive insects (balsam woolly adelgid), ozone pollution (worst in the eastern United States), and increased temperatures associated with global warming. Temperature change, in particular, will have a significant long-term effect on the Lepidoptera species in the Park. High-elevation forests could be eliminated slowly as climate changes, taking many of the endemic and disjunct species with them. There is some evidence that at least one species already has been extirpated from the Park. Polygonia faunus was recorded regularly from the early days of the Park through the 1950s. Few records exist after that, and none since 1983, despite frequent visitation by lepidopterists and other entomologists. Lepidoptera diversity in GSMNP is tied to vascular plant diversity, which is due to the wide range of habitats and physical features, such as slope, soil type, elevation, and the associated temperature and moisture conditions. Any threat to plant diversity or unique habitats will have an effect on the diversity of Lepidoptera. It would be worthwhile to monitor populations of endemic, disjunct, and specialist species along with the host plants and habitats that support them. DNA sampling and cryopreservation of specimens provide a resource for current and future systematists. With the expertise assembled at a bioblitz, 204 Southeastern Naturalist Special Issue 1 we can sort and identify a large number of species. With fresh specimens, we get 100% DNA recovery and greatly reduce the cost per individual specimen. COI sequences obtained from these specimens will serve not only as a potential identification barcode, but also as a significant piece of data for specialists as they estimate phylogenies and revise groups. Cryopreservation of specimens allows researchers to recover other mitochondrial and nuclear genes as well as other tissues as needed for future studies. A goal of future inventories should be to have as large a percentage of the fauna barcoded and cryopreserved as possible. Any attempt to comprehensively catalogue the species of a state, region, or locality will be a multi-year effort and, because of transient taxa and faunal change, can never be complete. After seven years of survey effort in GSMNP, we now have a nearly complete list of all resident Macrolepidoptera species. We have recorded 1047 species of Macrolepidoptera and 795 species of Microlepidoptera (Table 2). The four most species-rich groups in the Park are the Noctuidae s. l. (Noctuidae + Nolidae + Pantheidae, 549 spp.); Geometridae (225 spp.); Pyralidae s. l. (Pyralidae + Crambidae, 221 spp.); and Tortricidae (218 spp.) (see Table 1 for family totals). For well-surveyed groups, only a few additional resident species can be expected. Continued sampling undoubtedly will record additional transient species. In addition, new species are expected to become established in the Park (as well as others lost) as a result of long-term faunal changes, many of which are expected to result from global warming. Our sampling effort and results compare favorably with other relatively complete surveys of Lepidoptera (Proctor [1946] for Mount Desert Island, D.L.Wagner (unpubl. data) for Connecticut, Forbes [1923–60] for New York). For example, the current survey for the state of Connecticut has a comparable total species list (but what is thought to be a nearly complete Microlepidoptera list), although the area surveyed is about 10 times as large as GSMNP. The increased emphasis on leafmine surveys and larval sampling of the Connecticut fauna has contributed significantly to its more complete Microlepidoptera list, and the GSMNP survey will require this sort of effort to inventory more completely the Microlepidoptera in the Park. A major step in this direction will be a focused effort to collect and identify species in the Gelechioidea, the largest under-sampled Lepidoptera taxon in GSMNP. The information gained from the Lepidoptera survey of GSMNP is now being disseminated in a variety of formats. Much of the fieldwork for Wagner’s (2005) field guide to caterpillars in North America was conducted over a three-year period in the Park with grants from Discover Life in America. Pogue (2005, 2006) continues to work on a series of papers detailing the occurrence and distribution of noctuid species in the Park. A current list of all species known from the Park is in preparation, and several other papers on the Park’s lepidopteran fauna are in various stages of completion. The data analyzed in these papers will provide GSMNP with a very powerful tool for monitoring and managing the biota of the Park and be a basis for research in 2007 B.G. Scholtens and D.L. Wagner 205 many areas of Lepidoptera biology, and the methods developed for the ATBI will serve as a model for future surveys. Acknowledgments We owe debts of gratitude to many individuals and organizations. Many researchers, park personnel, and volunteers have been involved in the survey. James Adams, Tommy Allen, Bill Black, John Brown, Richard Brown, Sibyl Bucheli, Michael Canfield, Cameron Cheri, Chuck Cooper, Charlie Covell, Don Davis, Jeremy deWaard, Mark Epstein, Irving Finkelstein, Loran Gibson, Paul Goldstein, James Hayden, Paul Hebert, Jeanie Hilton, John Himmelman, Ronald Hodges, Eric Hossler, Marcia Jumblatt, Leroy Koehn, Jean-François Landry, Sangmi Lee, Nancy Lowe, Edda Martinez, Mark Mello, Eric Metzler, Michael Pogue, Jerry Powell, Michelle Prysby, Bob Pyle, Jadranka Rota, Dale Schweitzer, Craig Segebarth, Ian Segebarth, Kevin Segebarth, Tom Smith, Ian and Stephanie Stocks, Bo Sullivan, Geoff Tucker, Rene Twarkins, Henning von Schmeling, Monty Volovski, Jessica Watson, and Don Wright all participated in at least one bioblitz effort. Participating Park and USGS staff included Don Defoe, Keith Langdon, Adriean Mayer, Becky Nichols, Chuck Parker, and Paul Super. The instructors and students at both the Tremont Environmental Education Center and the Appalachian Highlands Learning Center have continuously provided samples since the beginning of the ATBI. James Adams helped formulate our list of significant species. Jim Froyd constructed a mobile shelf unit that has proved especially valuable during the bioblitzes. Primary funding for survey efforts has come from multiple grants from Discover Life in America to the authors, Michael Pogue, John Brown and Don Davis, and Eric Metzler. DLIA also provided lodging, logistical assistance, and coordinated volunteers and TWiG outreach efforts. Support for D.L. Wagner’s efforts also came from the US Department of Agriculture, Forest Services, Forest Health Technology Enterprise Team, cooperative agreement number 01-CA- 11244225-215. In addition, support has come from the Smithsonian Institution, grants to Paul Hebert and many in-kind donations from researchers. The manuscript was improved by comments from James Adams and other reviewers. Literature Cited Adams, J.K. 2002. The moth fauna of the Highlands area and possible impacts from 1999 btk spraying: Sampling the non-target moth fauna in 2001. Report to USDA Forest Service, Nantahala National Forest, Highlands, NC. 41 pp. Braun, A.F. 1935. Notes and new species of Microlepidoptera. Transactions of the American Entomological Society 61:45–52. Brower, A.E. 1974–1986. A list of the Lepidoptera of Maine. Parts 1, 2. Maine Agricultural Experiment Station Technical Bulletin 66:1–136; 109:1–60; 114:1–70. Brown, J.W., and K. Bash. 1997 (2000). The Lepidoptera of Marine Corps Air Station Miramar: Calculating faunal similarity among sampling sites and estimating total species richness. Journal of Research on the Lepidoptera 36:45–78. Brown, R.L. 1986. Survey of arthropods in southern Appalachians - Great Smoky Mountains National Park. Report to the National Park Service, Great Smoky Mountains National Park, Gatlinburg, TN. 13 pp. Forbes, W.T.M. 1923–1960. Lepidoptera of New York and neighboring states, I–IV. Cornell University Agricultural Experiment Station Memoirs 68:1–729; 274:1– 263; 329:1–433; 371:1–188. 206 Southeastern Naturalist Special Issue 1 Hebert, P.D.N., A. Cywinska, S.L. Ball, and J.R. deWaard. 2003. Biological identifications through DNA barcodes. Proceedings of the Royal Society B 270:313–321. Hebert, P.D.N., E.H. Penton, J.M. Burns, D.H. Janzen, and W. Hallwachs. 2004. Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfl y Astraptes fulgerator. Proceedings of the National Academy of Sciences 101:14812–14817. Hodges, R.W., T. Dominick, D.R. Davis, D.C. Ferguson, J.G. Franclemont, E.G. Munroe, and J.A. Powell (Eds.). 1983. Check List of the Lepidoptera of America North of Mexico. E.W. Classey Limited and The Wedge Entomological Research Foundation, London. UK and Washington, DC. 284 pp. Janzen, D.H., M. Hajibabaei, J.M. Burns, W. Hallwachs, E. Remigio, and P.D.N Hebert. 2005. Wedding biodiversity inventory of a large and complex Lepidoptera fauna with DNA barcoding. Philosophical Transactions of the Royal Society B 360:1835–1845. Kristensen, P.N. (Ed.). 1999. Lepidoptera, Moths, and Butterfl ies. Volume 1: Evolution, Systematics, and Biogeography. Handbook of Zoology Volume IV Arthropods: Insecta Part 35. Walter de Gruyter, Berlin, Germany, and New York, NY. 491 pp. Lambdin, P.L., D. Paulsen, J.K. Watson, and K. Langdon. 1991. Checklist of the Lepidoptera in the Great Smoky Mountains National Park. University of Tennessee Agricultural Experiment Station Research Report 91–24:1–30. Metzler, E.H. 1999. Inventory of lepidopterans and other insects at nine sites in Great Smoky Mountains National Park, October 1999. Report to the National Park Service. 25 pp. Pogue, M.G. 2005. The Plusiinae (Lepidoptera: Noctuidae) of Great Smoky Mountains National Park. Zootaxa 1032:1–28. Pogue, M.G. 2006. The Noctuiinae (Lepidoptera: Noctuidae) of Great Smoky Mountains National Park. Zootaxa 1215:1–95. Pollard, E., and T.J. Yates. 1994. Monitoring Butterfl ies for Ecology and Conservation. Chapman and Hall, London, UK. 274 pp. Powell, J.A. 1995. Lepidoptera inventories in the continental United States. Pp. 168–170, In E. LaRoe, G. Farris, C. Puckett, P. Doran and M. Mac (Eds.). Our Living Resources. US Department of the Interior, Washington, DC. Proctor, W. 1946. Biological Survey of the Mount Desert Region. Part VII. The Insect Fauna. Wistar Institute of Anatomy and Biology, Philadelphia, PA. 566 pp. Sullivan, J.B. 2004. Lepidopteran studies: The North Shore site. Report to the National Park Service, Great Smoky Mountains National Park, Gatlinburg, TN. 6 pp. Sullivan, J.B., and L.C. Deutschmann. 2002. Lepidopteran studies: The Ravensford site. Report to the National Park Service, Great Smoky Mountains National Park, Gatlinburg, TN. 34 pp. Wagner, D.L. 2005. Caterpillars of Eastern North America. Princeton University Press, Princeton, NJ. 512 pp.