Influence of Imidacloprid and Horticultural Oil on Spider Abundance on Eastern Hemlock in the Southern Appalachians

Influence of Imidacloprid and Horticultural Oil on Spider Abundance on Eastern Hemlock in the... Abstract Hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae), is an exotic pest of eastern hemlock, Tsuga canadensis (L.) Carrière (Pinales: Pinaceae), in the eastern United States. Two commonly used insecticides to manage adelgid are imidacloprid, a systemic neonicotinoid insecticide, and horticultural oil, a refined petroleum oil foliar spray. We have investigated the influence of imidacloprid and horticultural oil on spider abundance at different canopy strata in eastern hemlock. In total, 2,084 spiders representing 11 families were collected from the canopies of eastern hemlock. In beat-sheet and direct observation samples, the families Theridiidae, Araneidae, Salticidae, and Anyphaenidae were the most abundant. Significantly higher numbers of spiders were recorded on untreated control trees compared with trees treated with imidacloprid using soil drench and soil injection applications. Spider abundance in trees injected with imidacloprid and horticultural oil applications did not significantly differ from control trees. Spider abundance was significantly greater in the top and middle strata of the canopy than in the bottom stratum, where imidacloprid concentrations were the highest. Regression analysis showed that spider abundance was inversely associated with imidacloprid concentration. This research demonstrates that imidacloprid, when applied with selected methods, has the potential to result in reductions of spider densities at different strata. However, slight reductions in spider abundance may be an acceptable short-term ecological impact compared with the loss of an untreated hemlock and all the associated ecological benefits that it provides. Future studies should include investigations of long-term impact of imidacloprid on spiders associated with eastern hemlock. spider, imidacloprid, eastern hemlock, Tsuga canadensis, nontarget Eastern hemlock, Tsuga canadensis (L.) Carrière, is a coniferous tree species native to the eastern United States and adjacent Canadian provinces (Welch and Haddow 1993), where it is an important economic, ecological, and aesthetic component of forest ecosystems (Quimby 1996, Brooks 2001). Hemlocks are long-lived and function as ‘foundation species’, and no other conifer tree species is capable of functionally replacing hemlock in eastern forests (Kizlinski et al. 2002, Orwig et al. 2002, Ellison et al. 2005, Orwig et al. 2008). Populations of eastern hemlock and Carolina hemlock, Tsuga caroliniana Engelmann, have dramatically declined in the eastern United States due to feeding by an invasive insect pest, hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae). All sizes and ages of eastern and Carolina hemlock are susceptible to this adelgid (McClure 1987, Eschtruth et al. 2006). Heavy infestations of hemlock woolly adelgid cause premature foliage drop and bud abortion, and death often occurs within as few as 4 yr after infestation (McClure 1991, McClure et al. 2001, Ward et al. 2004). Besides hemlock woolly adelgid, several other arthropod pests and plant pathogens attack eastern hemlock (Godman and Lancaster 1990, Souto and Shields 2000, Raupp et al. 2004, Ward et al. 2004). These include elongate hemlock scale, Fiorinia externa Ferris (Hemiptera: Diaspididae), hemlock looper, Lambdina fiscellaria fiscellaria (Guenèe) (Lepidoptera: Geometridae), spruce spider mite, Oligonychus ununguis (Jacobi) (Acari: Tetranychidae), hemlock borer, Melanophila fulvogutta (Harris) (Coleoptera: Buprestidae), hemlock rust mite, Nalepella tsugifoliae Keifer (Acari: Eriophyidae), and the pathogen that causes root rot disease, Armillaria mellea (Vahl: Fries) Kummer (Agaricales: Physalacriacaea). Mortality of hemlock in hemlock-dominated forests affects forest structure and composition, as well as ecosystem function. Forest canopies contain highly diversified arthropod species, due to structural heterogeneity and variations in resources (Wu et al. 1989, Schowalter and Ganio 1998). The canopy of eastern hemlock can be expansive and provides suitable habitat for numerous arthropod species. Several studies have investigated the arthropods associated with eastern hemlock, and more than 300 species of insects have been found in hemlock stands in the southern Appalachians (Wallace and Hain 2000, Buck et al. 2005, Lynch et al. 2006, Dilling et al. 2007, Dilling et al. 2009, Jetton et al. 2009). Eastern hemlock also may provide important habitat for arboreal spiders. Spiders are important predators in many temperate forest systems, and their abundance depends mainly upon habitat structure and prey abundance (Halaj et al. 2000). In Oregon, spiders comprised over 50% of the arthropod predators in old growth stands of western hemlock, Tsuga heterophylla (Rafinesque) Sargent (Pinales: Pinaceae) (Schowalter 1995). Additionally, spiders (n = 197 specimens) were among the most abundant arthropod orders collected on eastern hemlocks sampled in the Great Smoky Mountains National Park, comprising 21.7% of all arthropods collected (n = 906 specimens; Falcone and DeWald 2010). In a study comparing spider communities in the canopies of hemlock and deciduous tree species, more spider families and specimens were collected from eastern hemlock (n = 20 families and 3,706 specimens) than from deciduous canopies (n = 15 families and 324 specimens; Mallis and Rieske 2011). These studies, in conjunction with the documentation of insect species that occur in hemlock canopies, demonstrate the diversity of arthropods found in eastern hemlock canopies, and these species may be adversely impacted by loss of hemlock. Several chemical insecticides, including horticultural oil (a refined petroleum oil foliar spray), insecticidal soap, and imidacloprid (a neonicotinoid insecticide), play an important role in suppressing hemlock woolly adelgid and maintaining healthy hemlock trees (Hale 2004, Ward et al. 2004, Dilling et al. 2010, Mayfield et al. 2015, Benton et al. 2016). Although imidacloprid is used widely to manage hemlock woolly adelgid, little is known of the nontarget impacts of this pesticide on predaceous arthropods, especially spiders, in hemlock canopies. Laboratory-based bioassays conducted on two introduced predatory beetle species [Laricobius nigrinus Fender (Coleoptera: Derodontidae) and Sasajiscymnus tsugae (Sasaji and McClure) (Coleoptera: Coccinellidae)] that have been widely released in the eastern United States (Havill et al. 2014) documented lethal effects of imidacloprid, by both ingestion of treated adelgids and direct external contact, as well as sublethal intoxication (Eisenback et al. 2010). Limited information exists on the effect of imidacloprid on introduced or native predators in field settings despite laboratory assays. In one study in the Great Smoky Mountains National Park, no significant differences were observed in arthropod species composition in hemlock stands treated with imidacloprid compared with untreated stands 3–18 mo after treatments (Falcone and DeWald 2010). Also, Mayfield et al. (2015) found no residues of imidacloprid on L. nigrinus collected from imidacloprid-treated trees and no differences in abundance of Laricobius spp. larvae between treated and untreated trees 7 yr following treatment. However, if the abundance of predatory species are reduced on imidacloprid-treated hemlock trees, a greater abundance of some pest species may result. For example, abundance of hemlock rust mite and spruce spider mite was greater on hemlock trees treated with imidacloprid than on untreated trees, which was attributed to the removal of several predators, including spider species, that feed on the mite pests (Raupp et al. 2004). Further investigations on the effect of imidacloprid on predatory species in hemlock canopies are warranted to clarify the impact of imidacloprid, if any, on spider populations. The management of hemlock woolly adelgid using pesticides is warranted, considering the ecological importance of hemlock as a foundation species in forests and its importance as habitat for arthropods. However, lack of information on nontarget effects of these pesticides on predatory arthropods, especially spiders, necessitates additional research to evaluate potential negative impacts. Additionally, imidacloprid movement and persistence throughout foliage in the canopy have been shown to vary greatly throughout different stratum (Dilling et al. 2010, Benton et al. 2015), thus having the potential of varying impacts based on spatial dynamics. The potential impact of these varying levels of imidacloprid on spider abundances in canopies of eastern hemlock is unknown. We have previously reported effects of chemical treatments on nonpredator guilds in the canopy (Dilling et al. 2009, 2010). In this study, we investigated the effect of imidacloprid on vertical distribution of spiders on eastern hemlock. The objective of this study was to determine the effect of insecticide treatments on spider populations at three strata of the canopy of eastern hemlock. Materials and Methods Study Sites This study was conducted at Indian Boundary Recreation Area in the southern part of Cherokee National Forest, Monroe County, Tennessee. To evaluate the vertical stratification of predators in hemlock canopies and determine the impact of insecticides on spiders, on 5 November 2005, eastern hemlocks (n = 30) were selected and arranged into three blocks with 10 trees each (35° 23.787 N, 84° 06. 662 W, elevation: 543 m; 35° 23.764 N, 84° 06.732 W, elevation: 555 m; 35° 24.173 N, 84° 06.268 W, elevation: 565 m). Trees in each block were arranged in five pairs, with one tree in each pair treated with one of five treatments (imidacloprid as tree injection, soil injection, and soil drench; horticultural oil as foliar spray; and untreated control) on one of two application times (29–30 November 2005 [Fall 2005] or 16 April 2006 [Spring 2006]). Tree pairs were assigned based on their similarity in height, condition, and adelgid infestation level. Blocks were located in a short-leaf pine-oak (Type 76) forest (Eyre 1980, Hakeem 2008). Insecticide Treatments The tree injection system consisted of the Mauget 3-ml 10% imicide capsules and feeder tubes (J. J. Mauget Co., Arcadia, CA). Imidacloprid was applied via tree injection at the rate of one capsule per 15 cm dbh (0.15 ml [AI]/2.5 cm dbh). A 1.2-cm hole was made ca. 20.5 cm above ground level using a 0.4-cm bit drilled at a slanting angle. The feeder tube was inserted into the hole, and a capsule was attached to the tube. Capsules were equally distributed on the tree trunk as much as possible for even distribution of insecticide. The capsules were left attached until uptake of the material was complete (ca. 1–5 h). Imidacloprid was applied as a soil injection using a Kioritz soil injector (Kioritz Corp., Tokyo, Japan). Merit 75 WP insecticide (Bayer, Kansas City, MO) was diluted to 1 g (AI)/2.5 cm dbh in 60 ml of water. Soil injection was applied within 45 cm of the base of the trunk and spaced as evenly as possible. The insecticide was injected at a depth of 7 cm below soil surface at 30 ml per injection. The soil drench treatment, applied using a high-pressure hydraulic sprayer (FMC Corporation, Jonesboro, AR), consisted of imidacloprid (Merit 75 WP) at the rate of 1.5 g (AI)/2.5 cm dbh. The standard recommended dose of 50 g of imidacloprid was mixed with 379-liter water. The soil beneath the drip line of the canopy was sprayed with 125 liters of mixture to provide maximum coverage. SunSpray horticultural oil (Sun Co., Philadelphia, PA) was applied as a foliar spray using a high-pressure hydraulic sprayer (FMC Corporation). The horticultural oil foliar spray was applied at the rate of 7.6 liters (AI)/379 liters of water and sprayed on the canopy until runoff to ensure that the tree was fully covered. Sampling of Vertical Strata Sampling of spiders in the canopy of eastern hemlock was conducted twice: 16–19 August 2006 and 5–11 September 2007. Each tree canopy was divided into three equal vertical sections, or strata, by height: top (upper 1/3 of canopy), middle (middle 1/3 of canopy), and bottom (lower 1/3 of canopy). A bucket truck (Genie Z 45/22, Tigard, OR) was used to collect samples from the three strata of each tree. Beat-sheet sampling and handpicking were the collection methods used to collect spiders from the canopy. From each stratum, three random beat-sheet samples (0.7 × 0.7 m; Bioquip Products, Rancho Dominguez, CA) were taken by holding the beat sheet underneath the branch and striking the branch five times with a wooden rod (1 m). Handpicking specimens involved collecting visually observed specimens for 5 min per stratum using forceps or fingers. Specimens collected during both sampling methods were placed in vials containing 75% alcohol and taken to the laboratory for processing and identification. Spiders were identified, where possible, by the author (AH) to family using Howel and Jenkins (2004) and Ubick et al. (2005), and family identifications were confirmed by Pat Barnwell, University of Tennessee. The specimens are housed in the Integrated Pest Management and Biological Control Laboratory, the University of Tennessee, Knoxville. Quantification of Imidacloprid Imidacloprid concentrations in twig and needle tissue were determined in a synchronous study by Dilling et al. (2010). Commercially available enzyme-linked immunosorbent assay kits were used to measure the imidacloprid residues within branches to determine the potential effect of imidacloprid on spiders (EnviroLogix 2015). In this test, the compound horseradish peroxidase-labeled imidacloprid was used, which competes with the imidacloprid residues present in the sample for a limited number of antibody sites on the walls of the test wells. This kit was used to quantify concentrations of imidacloprid between 0.2 and 6.0 ppb. Sample preparation and analysis are detailed in Dilling et al. (2010). Data Analysis Data were analyzed as a randomized block design with two fixed factors (treatment and application time) and two repeated measures (sampling year and strata). Specimens collected from beat-sheet sampling and handpicking were pooled to represent spider abundance per stratum per tree per sampling year. Treatment, application time, stratum, and sampling year were considered independent variables, and spider abundance per stratum per tree per sampling year was used as the response variable. All interactions among these variables were analyzed for significant associations. Shapiro–Wilk W test for normality and Levene’s test of homogeneity of variance were used to verify that data conformed to the assumptions of analysis of variance (ANOVA). Due to heterogeneity of group variances, data were rank transformed, and all data were analyzed by one-way ANOVA (ANOVA; PROC MIXED). Significance for all ANOVA tests was set at P ≤ 0.05. When significant ANOVA results were obtained, post hoc analysis to detect differences among means was determined using Tukey–Kramer test (α = 0.05). Kenward-–Roger method was used to adjust degrees of freedom for repeated measures because repeated measurements are correlated and consequently require an adjustment in the error degrees of freedom used in calculating significance tests. Original means and standard error values are reported. Similarity in family diversity and abundance among samples representing community composition was determined using analysis of similarity (ANOSIM) on Bray–Curtis distances constructed in a resemblance matrix using PRIMER version 6 software (Clarke and Warwick 1997) across tree strata and treatment type. Differences in samples were plotted using nonmetric multidimensional scaling. The relationship between imidacloprid concentrations at each stratum and mean spider specimen abundance was analyzed with linear regression analysis using PROC REG. Mean imidacloprid concentrations in parts per billion (ppb) determined from composite twig and branch samples presented in Dilling et al. (2010) that coincide with the vertical sampling dates presented in this study were regressed with the respective mean spider abundance for each stratum, treatment, and application time. All data analyses were conducted using SAS/STAT software, version 9.1 of the SAS System for Windows (SAS Institute 2006). Results and Discussion Density and Diversity of Spiders A diverse group of spiders was recovered from canopies of eastern hemlock. In total, 2,091 spiders were collected from canopies of eastern hemlock during this study. Of these, 2,084 spiders belong to 11 families; the families of 7 individuals could not be determined (Table 1). In general, spiders were distributed throughout the tree strata. Higher numbers (n = 1,588) of spiders were collected in beat-sheet samples compared with direct observations (n = 503), and greater numbers of spiders were collected in 2006 (n = 1,144) than in 2007 (n = 947). The most abundant families were Theridiidae, Araneidae, Salticidae, and Anyphaenidae. These four families represented 83.5% of all spiders collected (Table 1). Table 1. Spider families recorded from different strata on eastern hemlock, Indian Boundary Campground, Cherokee National Forest, Tennessee, 2006 and 2007 Spider family Stratum* Treatment† Total abundance‡ 2006 2007 2006 2007 Araneidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 445 Anyphaenidae B, M, T B CO, HO, SD, SI, TI CO, HO, SD, SI, TI 319 Corinnidae M SI, TI 2 Clubionidae B, T CO, SD, SI, TI HO, TI 16 Dictynidae B, T B, M CO, SD, SI, TI TI, SI 6 Linyphiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Salticidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 374 Tetragnathidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Theridiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 609 Thomisidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 186 Uloboridae B, T B, M CO, SD, SI, TI HO, TI 13 Unknown B, T B, T CO, HO, SD, SI, TI CO, TI 7 Spider family Stratum* Treatment† Total abundance‡ 2006 2007 2006 2007 Araneidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 445 Anyphaenidae B, M, T B CO, HO, SD, SI, TI CO, HO, SD, SI, TI 319 Corinnidae M SI, TI 2 Clubionidae B, T CO, SD, SI, TI HO, TI 16 Dictynidae B, T B, M CO, SD, SI, TI TI, SI 6 Linyphiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Salticidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 374 Tetragnathidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Theridiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 609 Thomisidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 186 Uloboridae B, T B, M CO, SD, SI, TI HO, TI 13 Unknown B, T B, T CO, HO, SD, SI, TI CO, TI 7 * Strata from which spider families were collected: B = bottom; M = middle; T = top. † Treatments from which spider families were collected: CO = control; HO = horticultural oil; SD = soil drench; SI = soil injection; TI = tree injection. ‡ Total abundance represents the number of all individuals collected in each family. View Large Table 1. Spider families recorded from different strata on eastern hemlock, Indian Boundary Campground, Cherokee National Forest, Tennessee, 2006 and 2007 Spider family Stratum* Treatment† Total abundance‡ 2006 2007 2006 2007 Araneidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 445 Anyphaenidae B, M, T B CO, HO, SD, SI, TI CO, HO, SD, SI, TI 319 Corinnidae M SI, TI 2 Clubionidae B, T CO, SD, SI, TI HO, TI 16 Dictynidae B, T B, M CO, SD, SI, TI TI, SI 6 Linyphiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Salticidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 374 Tetragnathidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Theridiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 609 Thomisidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 186 Uloboridae B, T B, M CO, SD, SI, TI HO, TI 13 Unknown B, T B, T CO, HO, SD, SI, TI CO, TI 7 Spider family Stratum* Treatment† Total abundance‡ 2006 2007 2006 2007 Araneidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 445 Anyphaenidae B, M, T B CO, HO, SD, SI, TI CO, HO, SD, SI, TI 319 Corinnidae M SI, TI 2 Clubionidae B, T CO, SD, SI, TI HO, TI 16 Dictynidae B, T B, M CO, SD, SI, TI TI, SI 6 Linyphiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Salticidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 374 Tetragnathidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Theridiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 609 Thomisidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 186 Uloboridae B, T B, M CO, SD, SI, TI HO, TI 13 Unknown B, T B, T CO, HO, SD, SI, TI CO, TI 7 * Strata from which spider families were collected: B = bottom; M = middle; T = top. † Treatments from which spider families were collected: CO = control; HO = horticultural oil; SD = soil drench; SI = soil injection; TI = tree injection. ‡ Total abundance represents the number of all individuals collected in each family. View Large The number of specimens and spider families in this study are similar to those observed in other studies of spiders associated with eastern hemlock. In mature stands of eastern hemlock in the Great Smoky Mountains National Park, more spider families (n = 16) but fewer specimens (n = 191) were collected from pitfall traps (Jetton et al. 2009). However, limited direct comparison can be made between this study and the current study, as spider families and specimen abundance in terrestrial systems could vary greatly from what is present in canopies. Greater spider specimen abundance was observed in the current study compared with that observed in hemlock canopies in another study in the Great Smoky Mountains National Park (n = 197; Falcone and DeWald 2010). However, family identifications were not performed for spiders collected in hemlock canopies in the Smokies (Falcone and DeWald 2010), and family numbers cannot be compared to this study. In a study comparing spider populations of hemlock canopies versus deciduous canopies in Kentucky, more specimens (n = 3,706) and families (n = 20) were collected from untreated hemlock canopies than in the current study (Mallis and Rieske 2011). The number of samples (24 vs 2 samples) may have been the greatest influence on the differences in spiders observed in Kentucky compared to this study. Sampling was conducted monthly over a 24-mo period in Kentucky (Mallis and Rieske 2011), whereas intensive spider sampling was conducted two times during this study period. The hemlock canopies evaluated as part of the current study contained an abundance of prey items (293 insect species; Dilling et al. 2009), with similar orders (Diptera, Psocoptera, Hymenoptera, and Hemiptera) comprising the bulk of prey items observed in Kentucky (Mallis and Rieske 2010). Therefore, while more spider families and specimens were observed by Mallis and Rieske (2011), family numbers and specimen abundance similar to those observed in Kentucky are expected from these study trees if sampling were conducted at similar frequencies. Vertical Distribution of Spider Families on Eastern Hemlock Community composition of families was not significantly different among tree strata (R = 0.054, n = 30, P > 0.001 or treatment types (R = 0.134, n = 30, P > 0.001). The similarity of family community composition across bottom, middle, and top strata indicates a homogenous distribution throughout the tree canopy. The similarity of family community composition across treatment types indicates a negligible impact on the community composition of spiders associated with eastern hemlock. This homogenous distribution could be due to vertical distribution of prey. Although community composition of family was similar among strata, the abundance of individuals within families, which will be discussed later, was distributed differently among strata. Impact of Insecticidal Treatments on Spiders Insecticidal treatments influenced the numbers of spiders collected from eastern hemlock. Significant differences (F4,19.9 = 5.73; P < 0.0031) in mean spider abundance per treatment per tree were observed among control (no treatments), soil drench, and soil injection (Table 2). In both sampling years, mean spider abundance in soil drench and soil injection treatments was significantly lower than control treatments, but no differences in spider abundance were observed among control trees, tree injection, and horticultural oil treatments (Table 2). Specimen abundance in samples collected in 2006 ranged from 19.88 spiders per tree in control treatments to 7.17 spiders per tree in soil drench treatments. Specimen abundance in samples collected in 2007 ranged from 17.41 spiders per tree in control treatments to 6.76 spiders per tree in soil injection treatments. Spider abundance was numerically greater for all treatments in 2006 than in 2007, with the exception of soil drench treatments, which were numerically greater in 2007 than in 2006. Spider abundance in soil injection and soil drench treatments also was significantly lower than control treatments for the overall mean of both years combined, with overall spider abundance per tree ranging from 18.64 in control treatments to 3.79 in soil injection treatments. However, no differences in spider abundance were observed among control trees, tree injection, and horticultural oil treatments for either sampling year or the overall mean of both years (Table 2). Additionally, in both sampling years and for the overall mean of both years combined, no significant differences were observed within imidacloprid treatments (tree injection, soil injection, and soil drench). No significant interactions were observed among treatment, application time, sampling year, stratum, or all. Table 2. Comparison of spider abundance (mean ± SE) on eastern hemlock, Tsuga canadensis, treated with different insecticides, 2006 and 2007 Treatment Year Average per year 2006* 2007 Control (not treated) 19.88 ± 2.32aA 17.41 ± 1.74aA 18.64 ± 2.03aA Tree injection 15.33 ± 1.85abA 13.83 ± 2.58abA 14.58 ± 2.21abA Horticultural oil 13.44 ± 2.57abA 10.63 ± 1.73abA 12.03 ± 2.15abA Soil drench 7.17 ± 0.78bA 7.80 ± 1.58bA 7.48 ± 1.18bA Soil injection 8.83 ± 1.01bA 6.76 ± 1.02bA 3.79 ± 1.01bA Treatment Year Average per year 2006* 2007 Control (not treated) 19.88 ± 2.32aA 17.41 ± 1.74aA 18.64 ± 2.03aA Tree injection 15.33 ± 1.85abA 13.83 ± 2.58abA 14.58 ± 2.21abA Horticultural oil 13.44 ± 2.57abA 10.63 ± 1.73abA 12.03 ± 2.15abA Soil drench 7.17 ± 0.78bA 7.80 ± 1.58bA 7.48 ± 1.18bA Soil injection 8.83 ± 1.01bA 6.76 ± 1.02bA 3.79 ± 1.01bA *Means followed by the same lowercase letter(s) within the same column, and means followed by the same uppercase letter within the same row are not significantly different (P ≤ 0.05; Tukey–Kramer test). View Large Table 2. Comparison of spider abundance (mean ± SE) on eastern hemlock, Tsuga canadensis, treated with different insecticides, 2006 and 2007 Treatment Year Average per year 2006* 2007 Control (not treated) 19.88 ± 2.32aA 17.41 ± 1.74aA 18.64 ± 2.03aA Tree injection 15.33 ± 1.85abA 13.83 ± 2.58abA 14.58 ± 2.21abA Horticultural oil 13.44 ± 2.57abA 10.63 ± 1.73abA 12.03 ± 2.15abA Soil drench 7.17 ± 0.78bA 7.80 ± 1.58bA 7.48 ± 1.18bA Soil injection 8.83 ± 1.01bA 6.76 ± 1.02bA 3.79 ± 1.01bA Treatment Year Average per year 2006* 2007 Control (not treated) 19.88 ± 2.32aA 17.41 ± 1.74aA 18.64 ± 2.03aA Tree injection 15.33 ± 1.85abA 13.83 ± 2.58abA 14.58 ± 2.21abA Horticultural oil 13.44 ± 2.57abA 10.63 ± 1.73abA 12.03 ± 2.15abA Soil drench 7.17 ± 0.78bA 7.80 ± 1.58bA 7.48 ± 1.18bA Soil injection 8.83 ± 1.01bA 6.76 ± 1.02bA 3.79 ± 1.01bA *Means followed by the same lowercase letter(s) within the same column, and means followed by the same uppercase letter within the same row are not significantly different (P ≤ 0.05; Tukey–Kramer test). View Large These results indicate that selected methods of imidacloprid applications can cause a short-term reduction in spider populations. Furthermore, no differences observed between sampling years for all treatments, in conjunction with no significant interactions between application time and treatment, suggests that timing of applications has no short-term influence on spider abundance. During both sampling years, spiders were least abundant in trees treated with imidacloprid applied with a soil injection or soil drench compared with control, horticultural oil, and trunk injection treatments. These reductions in spider numbers are the first documentation of short-term, nontarget impacts of imidacloprid on spiders in eastern hemlock canopies. Although the long-term impacts of imidacloprid on spider populations are unknown, these populations are expected to rebound over time as imidacloprid concentrations lessen in the plant tissues. These results differ from those reported spider populations in the Great Smoky Mountains National Park, which showed no differences in spider densities between treated and untreated hemlock trees (Falcone and DeWald 2010). In their study, they applied imidacloprid as a soil drench, while the current study had three imidacloprid application types and a foliar application of horticultural oil. The interval between application and sampling was similar in both studies, with the sampling interval ranging from ca. 1 mo (at one of six treated sites) to 18 mo in the Great Smoky Mountain National Park and from ca. 4 mo to ca. 22 mo in the current study. However, the total number of samples (n = 240) taken during the Great Smoky Mountains National Park study was less than the total number of samples in the current study (n = 720). In their study, spiders were sampled by branch clipping instead of beat-sheet sampling or handpicking, and branch clipping may not be an adequately intensive sampling method to properly assess spider abundance in forest settings, as it may allow some spiders the opportunity to flee during sampling. The combination of fewer samples and branch clipping as a sampling technique may have influenced the numbers of spiders collected in the Great Smoky Mountains National Park and prevent adequate comparisons to our findings. Trees treated with imidacloprid as a trunk injection consistently had greater numbers of spiders than horticultural oil, soil injection, or soil drench treatments. Trunk injection requires the trunk of trees to be drilled to allow insertion of the imidacloprid delivery apparatus. However, the lack of reductions of spiders on trees treated with imidacloprid applied as a trunk injection is unclear. Although no significant differences (F8,92.6 = 1.99; P = 0.0556) were observed among interactions of treatment, application time, and stratum, the trend of soil injection and soil drench treatments containing fewer spiders than other treatments is evident (Fig. 1). In fall and spring application times and all strata, soil injection and soil drench treatments were the lowest, with the exception of trees treated with horticultural oil in Spring 2006. For these trees, spider abundance was lower in horticultural oil treatments than soil injection, as well as tree injection and control treatments. The reason for this low level of spider abundance in trees treated with horticultural oil at one stratum for one application time is unclear. These nonsignificant results may be due to the complex nature of interactions between treatments, season, and strata, although treatments and strata were significant (Fig. 1). Fig. 1. View largeDownload slide Comparison of spider abundance (mean ± SE) within each treatment and strata for hemlock trees treated in Fall 2005 and Spring 2006. No significant differences (F8,92.6 = 1.99; P = 0.0556) were observed in interactions among treatment, application time, strata, or all. Fig. 1. View largeDownload slide Comparison of spider abundance (mean ± SE) within each treatment and strata for hemlock trees treated in Fall 2005 and Spring 2006. No significant differences (F8,92.6 = 1.99; P = 0.0556) were observed in interactions among treatment, application time, strata, or all. Canopy location (i.e., stratum) also influenced spider abundance. Significant differences (F2,92.6 = 5.60; P < 0.0051) were observed between mean spider abundance per sample per year between the top and the middle strata compared with the bottom stratum (Table 3). In both sampling years, significantly lower numbers of spiders were observed from bottom stratum than either the top or middle stratum, but no differences were observed in spider abundance in each respective stratum between years (Table 3). Specimen abundance in samples collected in 2006 ranged from 14.00 spiders per tree in the top stratum to 11.07 spiders per tree in the bottom stratum. Specimen abundance in samples collected in 2007 ranged from 11.56 spiders per tree in the top stratum to 8.56 spiders per tree in the bottom stratum. Significantly lower numbers of spiders also were recovered from the bottom stratum than from the top or middle stratum for the overall mean of both years combined, with spider abundance per tree ranging from 12.78 in the top stratum to 9.81 in the bottom stratum (Table 3). Higher numbers of spiders in the top strata is not surprising, as Joseph et al. (2011) also reported higher numbers of hemlock woolly adelgid ovisacs in the upper tree crown, suggesting less imidacloprid concentrations. No significant interactions were observed among stratum, treatment, application time, sampling year, or all. Oguri et al. (2014) compared spider assemblage between the upper and lower canopy layers (i.e., strata), and between the canopy and forest floor in evergreen cedar, Cryptomeria japonica D. Don (Pinales: Cupressaceae), and deciduous larch, Larix kaempferi [Lamb.] Carrieré (Pinales: Pinaceae), and found significant differences in number of species between strata in L. kaempferi only. Table 3. Comparison of spider abundance (mean ± SE) from three strata of eastern hemlock, Tsuga canadensis, 2006 and 2007 Strata Year Average per year 2006* 2007 Top 14.00 ± 1.95aA 11.56 ± 1.26aA 12.78 ± 1.60aA Middle 13.52 ± 1.57aA 13.72 ± 1.88aA 13.62 ± 1.72aA Bottom 11.07 ± 1.27bA 8.56 ± 1.40bA 9.81 ± 1.33bA Strata Year Average per year 2006* 2007 Top 14.00 ± 1.95aA 11.56 ± 1.26aA 12.78 ± 1.60aA Middle 13.52 ± 1.57aA 13.72 ± 1.88aA 13.62 ± 1.72aA Bottom 11.07 ± 1.27bA 8.56 ± 1.40bA 9.81 ± 1.33bA *Means followed by the same lowercase letter within the same column, and means followed by the same uppercase letter within the same row are not significantly different (P ≤ 0.05; Tukey–Kramer test). View Large Table 3. Comparison of spider abundance (mean ± SE) from three strata of eastern hemlock, Tsuga canadensis, 2006 and 2007 Strata Year Average per year 2006* 2007 Top 14.00 ± 1.95aA 11.56 ± 1.26aA 12.78 ± 1.60aA Middle 13.52 ± 1.57aA 13.72 ± 1.88aA 13.62 ± 1.72aA Bottom 11.07 ± 1.27bA 8.56 ± 1.40bA 9.81 ± 1.33bA Strata Year Average per year 2006* 2007 Top 14.00 ± 1.95aA 11.56 ± 1.26aA 12.78 ± 1.60aA Middle 13.52 ± 1.57aA 13.72 ± 1.88aA 13.62 ± 1.72aA Bottom 11.07 ± 1.27bA 8.56 ± 1.40bA 9.81 ± 1.33bA *Means followed by the same lowercase letter within the same column, and means followed by the same uppercase letter within the same row are not significantly different (P ≤ 0.05; Tukey–Kramer test). View Large Impact of Imidacloprid Concentrations on Spider Abundance Spider abundance in the canopy strata was directly associated with imidacloprid concentrations, as linear regression showed a strong negative association between mean specimen abundance across both application times and treatments and mean imidacloprid concentrations across all strata in both sampling years (Figs. 2 and 3). Different application methods of imidacloprid resulted in significant differences in imidacloprid concentrations distributed throughout the canopy, but for all imidacloprid application methods concentrations were highest in the bottom stratum and decreased in the middle and top strata (Dilling et al. 2010). In the top stratum, imidacloprid concentrations ranged from 0 to 98.44 ppb across all sampling times and treatments (Figs. 2A and D, 3A and D). Mean specimen abundance in the top stratum ranged from 1.62 to 5.19 per application time per sampling year per treatment per tree across all sampling times and treatments and declined as imidacloprid concentrations increased (Figs. 2A and D, 3A and D). In the middle stratum, imidacloprid concentrations ranged from 0 to 162.98 ppb across all sampling times (Figs. 2B and E, 3B and E). Mean specimen abundance in the middle stratum ranged from 0.48 to 5.15 per application time per sampling year per treatment per tree across all sampling times and treatments (Figs. 2B and E, 3B and E). Imidacloprid concentrations in the bottom stratum ranged from 0 to 199.33 ppb across all sampling times (Figs. 1C and F,2C and F). Mean specimen abundance in the bottom stratum ranged from 0.23 to 5.42 per application time per sampling year per treatment per tree across all sampling times and application types (Figs. 2C and F, 3C and F). Fig. 2. View largeDownload slide Relationship between mean spider abundance per tree based on the average number of specimens collected from trees treated in Fall 2005 within the top (A), middle (B), and bottom (C) strata, trees treated in Spring 2006 within the top (D), middle (E), and bottom (F) strata, and the mean imidacloprid concentrations (ppb) within each respective stratum for samples collected in August 2006. Fig. 2. View largeDownload slide Relationship between mean spider abundance per tree based on the average number of specimens collected from trees treated in Fall 2005 within the top (A), middle (B), and bottom (C) strata, trees treated in Spring 2006 within the top (D), middle (E), and bottom (F) strata, and the mean imidacloprid concentrations (ppb) within each respective stratum for samples collected in August 2006. Fig. 3. View largeDownload slide Relationship between mean spider abundance per tree based on the average number of specimens collected from trees treated in Fall 2005 within the top (A), middle (B), and bottom (C) strata, trees treated in Spring 2006 within the top (D), middle (E), and bottom (F) strata, and the mean imidacloprid concentrations (ppb) within each respective stratum for samples collected in September 2007. Fig. 3. View largeDownload slide Relationship between mean spider abundance per tree based on the average number of specimens collected from trees treated in Fall 2005 within the top (A), middle (B), and bottom (C) strata, trees treated in Spring 2006 within the top (D), middle (E), and bottom (F) strata, and the mean imidacloprid concentrations (ppb) within each respective stratum for samples collected in September 2007. These results indicate that the stratum in which spiders occur significantly influenced their abundance. Also, no differences observed between sampling years for all strata, in conjunction with no significant interactions between application time and strata, suggest that timing of applications do not influence short-term spider abundance in strata. During both sampling years and for the mean of both years combined, spiders were least abundant in the bottom stratum. Low numbers of spiders on the bottom stratum were directly associated with higher concentrations of imidacloprid in the bottom stratum, and high R2 values for all regressions demonstrate the strong associations between spider abundance within treatments and concentrations of imidacloprid at different strata. Therefore, certain methods of imidacloprid application (soil injection and soil drench) may not only reduce spider densities, but can have varying impacts on spiders throughout the canopy of eastern hemlock. The association between spider abundance and concentrations of imidacloprid observed during this study provides insight into what concentrations may influence spider abundance in the field. Imidacloprid concentrations of 120–250 ppb have been reported to suppress populations of hemlock woolly adelgid over time and cause lethal and sublethal effects to the adelgid predators L. nigrinus and S. tsugae (Tattar et al. 1998, Cowles et al. 2006, Eisenback et al. 2010). However, the accumulation of imidacloprid in prey items other than hemlock woolly adelgid that may cause lethal and sublethal effects in spiders is unknown. Regression analyses in the current study suggest that imidacloprid concentrations in twig and needle tissues >120 ppb may predict decreases in spider populations. For example, spider abundance on soil drenched trees treated in Spring 2006 and sampled in August 2006 (Fig. 2) was 2.20 spiders per tree at a concentration of 96.25 ppb in the top strata, yet only 0.62 spiders per tree at a concentration of 159.33 ppb in the middle strata. Furthermore, imidacloprid concentrations in all samples from the bottom stratum of soil injection- and soil drench-treated trees exceeded 120 ppb, with the lowest concentration (131.29 ppb) observed in soil injection-treated trees treated in Spring 2006 and sampled in September 2007 (Fig. 3). Imidacloprid concentrations do not exceed ca. 100 ppb in the top strata of the tree, and spider populations occurring in this stratum may not have experienced reductions during the study following imidacloprid treatments. Imidacloprid can translocate into the canopy quickly and have sustained effect on herbivores. Dilling et al. (2010) observed peaks in imidacloprid concentrations 9–12 mo following application, yet 3 mo following soil injection and soil drench applications concentrations in the bottom stratum of hemlocks were >169 ppb. Additionally, the bottom stratum maintained concentrations of imidacloprid >125 ppb 24 mo following soil injection and soil drench applications. Because imidacloprid can reach concentrations >120 ppb relatively quickly and sustain those concentrations in plant tissues over time, nontarget impacts to predatory species within the first 2 yr following imidacloprid application may be unavoidable. Nontarget impacts of imidacloprid on spider abundance are not surprising, as several studies have documented reduced spider abundance caused by applications of chemical pesticides in agricultural settings (Vickerman and Sunderland 1977, Salem and Matter 1991, Pekar 1999, Epstein et al. 2000, Meissle and Lang 2005, Cardenas et al. 2006, Markó et al. 2009). Furthermore, the presence and concentration of imidacloprid could influence spider numbers in several ways. First, direct contact with insecticides may be toxic to spiders and cause a reduction in spider abundance (Shaw et al. 2006). However, in the current study, due to the application methods used, it is unlikely that spiders had direct contact with imidacloprid. Second, imidacloprid could decrease numbers of herbivorous arthropod prey species, causing spiders to starve or leave the area in search of other prey. Dilling et al. (2009) observed significant reductions in specimen abundance in several insect guilds, including phytophagous insects, in trees where imidacloprid was applied as soil drench. These observed reductions could have increased competition among spiders, causing some to starve and/or cuing some spiders to leave in search of prey. Third, spiders may acquire imidacloprid in prey that have recently consumed treated plant material, causing secondary poisoning (Epstein et al. 2000). This secondary poisoning could lead to sublethal intoxication or death, depending on the concentration of the pesticide in the prey. Sublethal intoxication can cause reductions in adult longevity, fecundity, fertilization, and inhibition of feeding that can lead to decreases in specimen abundance (Devine et al. 1996, Boina et al. 2009). Decreases in spider abundance observed in the current study are most likely due to sublethal and/or lethal intoxication; however, more research is needed to determine the tritrophic movement of imidacloprid in this system. Spiders are generalist predators and feed on several pest species. Nontarget impact of pesticides that influence their abundance on hemlock may lead to further changes in arthropod communities associated with eastern hemlock. When considering applications of imidacloprid, dose, concentration, and application method should be considered to reduce nontarget impact on spiders. The findings reported herein demonstrate that imidacloprid applied as a soil injection or soil drench at the recommended rates can cause short-term reductions in spider populations in the lower canopy where imidacloprid concentrations are highest. Although the long-term effects of imidacloprid on spiders and other predatory arthropods are unknown, millions of untreated eastern hemlocks have died or are severely damaged, drastically altering hemlock and associated forest ecosystems. Imidacloprid provides an effective, economical treatment to protect and preserve some trees. Thus, slight reductions in spider abundance may be an acceptable short-term ecological impact compared with the loss of an untreated hemlock and all the associated ecological benefits that it provides. Monitoring of spider populations on these study trees over time, in conjunction with quantification of imidacloprid concentrations at the time of sampling, would document the persistence of imidacloprid in treated trees, as well as assess the impact of imidacloprid on spider population years after treatment. This information could inform conservation-oriented management strategies that are aimed at maintaining healthy canopies by reducing the impact of hemlock woolly adelgid, while simultaneously preserving diversity of predatory arthropods within the canopy of eastern hemlock. Acknowledgments We extend our special thanks to Joshua Grant, David Paulsen, Nicholas Reynolds, and Renee Follum (Department of Entomology and Plant Pathology, University of Tennessee, Knoxville, TN) for their assistance in collection of samples, Pat Barnwell for confirming family identifications, Ann Reed (Statistical Consulting Center, University of Tennessee, Knoxville, TN) for statistical assistance, and personnel at the Tellico Plains Ranger Station (Cherokee National Forest) for allowing us to conduct this study at Indian Boundary. This study was funded in part by USDA Forest Service (FS agreement no. 05-PA-11083150-070). References Cited Benton , E. P. , J. F. Grant , R. J. Webster , R. J. Nichols , R. S. Cowles , A. F. Lagalante , and C. I. Coots . 2015 . Assessment of imidacloprid and its metabolites in foliage of eastern hemlock multiple years following treatment for hemlock woolly adelgid, Adelges tsugae (Hemiptera: Adelgidae), in forested conditions . J. Econ. Entomol . 108 : 2672 – 2682 . Google Scholar CrossRef Search ADS PubMed Benton , E. P. , J. Grant , R. Cowles , J. Webster , R. Nichols , A. Lagalante , and C. Coots . 2016 . Assessing relationships between tree diameter and long-term persistence of imidacloprid and olefin to optimize imidacloprid treatments on eastern hemlock . Forest Ecol. Manag . 370 : 12 – 21 . Google Scholar CrossRef Search ADS Boina , D. R. , E. O. Onagbola , M. Salyani , and L. L. Stelinski . 2009 . Antifeedant and sublethal effects of imidacloprid on Asian citrus psyllid, Diaphorina citri . Pest Manag. Sci . 65 : 870 – 877 . Google Scholar CrossRef Search ADS PubMed Brooks , R. T . 2001 . Effects of the removal of overstory hemlock from hemlock-dominated forest on eastern redback salamanders . Forest Ecol. Manag . 149 : 197 – 204 . Google Scholar CrossRef Search ADS Buck , L. , P. Lambdin , D. Paulsen , J. Grant , and A. Saxton . 2005 . Checklist of insect species associated with eastern hemlock in the Great Smoky Mountains National Park and environs . Tenn. Acad. Sci . 80 : 1 – 10 . Cardenas , M. , F. Ruano , P. Garcia , F. Pascual , and M. Campos . 2006 . Impact of agricultural management on spider populations in the canopy of olive trees . Biol. Control 38 : 188 – 195 . Google Scholar CrossRef Search ADS Clarke , K. R. , and R. M. Warwick . 1997 . PRIMER user manual (Plymouth routines in multivariate ecological research) . Plymouth Marine Laboratory , Plymouth, UK . Cowles , R. S. , M. E. Montgomery , and C. A. S.-J. Cheah . 2006 . Activity and residues of imidacloprid applied to soil and tree trunks to control hemlock woolly adelgid (Hemiptera: Adelgidae) in forests . J. Econ. Entomol . 99 : 1259 – 1267 . Google Scholar CrossRef Search ADS Devine , G. J. , Z. K. Harlings , A. W. Scarr , and A. L. Devonshire . 1996 . Lethal and sublethal effects of imidacloprid on nicotine-tolerant Myzus nicotianae and Myzus persicae . Pestic. Sci . 48 : 57 – 62 . Google Scholar CrossRef Search ADS Dilling , C. , P. Lambdin , J. Grant , and L. Buck . 2007 . Insect guild structure associated with eastern hemlock in the southern Appalachians . Environ. Entomol . 36 : 1408 – 1414 . Google Scholar CrossRef Search ADS PubMed Dilling , C. , P. Lambdin , J. Grant , and R. Rhea . 2009 . Community response of insects associated with eastern hemlock to imidacloprid and horticultural oil treatments . Environ. Entomol . 38 : 53 – 66 . Google Scholar CrossRef Search ADS PubMed Dilling , C. , P. Lambdin , J. Grant , and R. Rhea . 2010 . Spatial and temporal distribution of imidacloprid in eastern hemlock in the southern Appalachians . J. Econ. Entomol . 103 : 368 – 373 . Google Scholar CrossRef Search ADS PubMed Eisenback , B. M. , S. M. Salom , L. T. Kok , and A. F. Lagalante . 2010 . Lethal and sublethal effects of imidacloprid on hemlock woolly adelgid (Hemiptera: Adelgidae) and two introduced predator species . J. Econ. Entomol . 103 : 1222 – 1234 . Google Scholar CrossRef Search ADS PubMed Ellison , A. M. , M. S. Bank , D. C. Barton , E. A. Colburn , K. Elliott , C. R. Ford , D. R. Foster , B. D. Kloeppel , J. D. Knoepp , G. M. Lovett , et al. 2005 . Loss of foundation species: consequences for the structure and dynamics of forested ecosystems . Front. Ecol. Environ . 3 : 479 – 486 . Google Scholar CrossRef Search ADS EnviroLogix . 2015 . Quantitative laboratory test for the detection of imidacloprid in water. Technical data sheet for Microwell Plate Assay, cat. No. EP006 . Retrieved April 25, 2018 from https://www.envirologix.com/wp-content/uploads/2015/05/EP006-Imidacloprid-101215.pdf. Epstein , D. L. , R. S. Zack , J. F. Brunner , L. Gut , and J. J. Brown . 2000 . Effects of broad-spectrum insecticides on epigeal arthropod biodiversity in Pacific Northwest apple orchards . Environ. Entomol . 29 : 340 – 348 . Google Scholar CrossRef Search ADS Eschtruth , A. K. , N. L. Cleavitt , J. J. Battles , R. A. Evans , and T. J. Fahey . 2006 . Vegetation dynamics in declining eastern hemlock stands: 9 years of forest response to hemlock woolly adelgid infestation . Can. J. Forest Res . 36 : 1435 – 1450 . Google Scholar CrossRef Search ADS Eyre , F . 1980 . Forest cover types of the United States and Canada . Society of American Foresters , Washington, DC . Falcone , J. F. , and L. E. DeWald . 2010 . Comparisons of arthropod and avian assemblages in insecticide-treated and untreated eastern hemlock (Tsuga canadensis [L.] Carr.) stands in Great Smoky Mountains National Park, USA . Forest Ecol. Manag . 260 : 856 – 863 . Google Scholar CrossRef Search ADS Godman , R. M. , and K. Lancaster . 1990 . Tsuga canadensis . In R. M. Burns and B. H. Honkala (eds.), Silvics of North America , vol. 1 . United States Department of Agriculture, Forest Service Hdbk. 654 , Washington, DC . Hakeem , A . 2008 . Non-target effect of imidacloprid on predatory guild on eastern hemlock, Tsuga canadensis (L.) Carriere, in the southern Appalachians . M.S. thesis, University of Tennessee , Knoxville , 78 pp. Halaj , J. , D. W. Ross , and A. R. Moldenke . 2000 . Importance of habitat structure to the arthropod food-web in douglas-fir canopies . Oikos . 90 : 139 – 152 . Google Scholar CrossRef Search ADS Hale , F. A . 2004 . The hemlock woolly adelgid: a threat to hemlock in Tennessee . The University of Tennessee, Agric. Ext. Serv ., Pub. SP503-G, Knoxville, TN . Havill , N. P. , L. C. Vieira , and S. M. Salom . 2014 . Biology and control of hemlock woolly adelgid . United States Department of Agriculture, Forest Service , Washington, DC . Howel , W. M. , and R. L. Jenkins . 2004 . Spiders of the eastern United States: a photographic guide . Pearson Education , New York, NY . Jetton , R. M. , F. L. Hasting , and F. P. Hain . 2009 . Arthropod diversity associated with old and secondary growth eastern hemlock (Tsuga canadensis) in the Great Smoky Mountains National Park . J. North Carolina Acad. Sci . 125 : 70 – 77 . Joseph , S. V. , J. L. Hanula , and S. K. Braman . 2011 . Distribution and abundance of Adelges tsugae (Hemiptera: Adelgidae) within hemlock trees . J. Econ. Entomol . 104 : 1918 – 1927 . Google Scholar CrossRef Search ADS PubMed Kizlinski , M. L. , D. A. Orwig , R. C. Cobb , and D. R. Foster . 2002 . Direct and indirect ecosystem consequences of an invasive pest on forests dominated by eastern hemlock . J. Biogeogr . 29 : 1489 – 1503 . Google Scholar CrossRef Search ADS Lynch , C. , P. Lambdin , J. Grant , R. Reardon , and R. Rhea . 2006 . Survey for established predators of elongate hemlock scale in Tennessee . Fla. Entomol . 89 : 527 – 528 . Google Scholar CrossRef Search ADS Mallis , R. E. , and L. K. Rieske . 2010 . Web orientation and prey resources for web-building spiders in eastern hemlock . Environ. Entomol . 39 : 1466 – 1472 . Google Scholar CrossRef Search ADS PubMed Mallis , R. E. , and L. K. Rieske . 2011 . Arboreal spiders in eastern hemlock . Environ. Entomol . 40 : 1378 – 1387 . Google Scholar CrossRef Search ADS PubMed Markó , V. , B. Keresztes , M. T. Fountain , and J. V. Cross . 2009 . Prey availability, pesticides and the abundance of orchard spider communities . Biol. Control 48 : 115 – 124 . Google Scholar CrossRef Search ADS Mayfield , A. , B. Reynolds , C. Coots , N. Havill , C. Brownie , A. Tait , J. Hanula , S. Joseph , and A. Galloway . 2015 . Establishment, hybridization and impact of Laricobius predators on insecticide-treated hemlocks: exploring integrated management of the hemlock woolly adelgid . Forest Ecol. Manag . 335 : 1 – 10 . Google Scholar CrossRef Search ADS McClure , M. S . 1987 . Biology and control of hemlock woolly adelgid . Connecticut Agricultural Experiment Station Bulletin 851 , New Haven, CT . McClure , M. S . 1991 . Density-dependent feedback and population cycles in Adelges tsugae (Homoptera: Adelgidae) on Tsuga canadensis . Environ. Entomol . 20 : 258 – 264 . Google Scholar CrossRef Search ADS McClure , M. S. , S. M. Salom , and K. S. Shields . 2001 . Hemlock woolly adelgid . United States Department of Agriculture, Forest Service, For. Health Technol. Ent. Team, FHTET-2001–03 , Morgantown, WV . Meissle , M. , and A. Lang . 2005 . Comparing methods to evaluate the effects of Bt maize and insecticide on spider assemblages . Agric. Ecosyst. Environ . 107 : 359 – 370 . Google Scholar CrossRef Search ADS Oguri , H. , T. Yoshida , A. Nakamura , M. Soga , and N. Hijii . 2014 . Vertical stratification of spider assemblages in two conifer plantations in central Japan . J. Arachnol . 42 : 34 – 43 . Google Scholar CrossRef Search ADS Orwig , D. A. , D. R. Foster , and D. L. Mausel . 2002 . Landscape patterns of hemlock decline in New England due to the introduced hemlock woolly adelgid . J. Biogeogr . 29 : 1475 – 1487 . Google Scholar CrossRef Search ADS Orwig , D. A. , R. C. Cobb , A. W. D’Amato , M. L. Kizlinski , and D. R. Foster . 2008 . Multi-year ecosystem response to hemlock woolly adelgid infestation in southern New England forests . Can. J. Forest Res . 38 : 834 – 843 . Google Scholar CrossRef Search ADS Pekar , S . 1999 . Effect of IPM practices and conventional spraying on spider population dynamics in an apple orchard . Agric. Ecosyst. Environ . 73 : 155 – 166 . Google Scholar CrossRef Search ADS Quimby , J. W . 1996 . Value and importance of hemlock ecosystems in the eastern United States , pp. 1 – 8 . In S. M. Salom , T. C. Tigner , and R. C. Reardon (eds.), Proceedings of the First Hemlock Woolly Adelgid Review , 12 October 1995, Charlottesville, VA . United States Department of Agriculture, Forest Service, For. Health Technol. Ent. Team, FHTET-96-10 , Morgantown, WV . Raupp , M. J. , R. E. Webb , A. Szczepaniec , D. Booth , and R. Ahern . 2004 . Incidence, abundance, and severity of mites on hemlocks following applications of imidacloprid . J. Arboric . 30 : 108 – 113 . Salem , S. A. , and M. M. Matter . 1991 . Relative effects of neem seed oil and deenate on the cotton leaf worm, Spodoptera littoralis Boisd, and the most prevalent predators in cotton fields at Menoufyia governorate . Bull. Fact Sci. Cairo Univ . 42 : 941 – 952 . SAS Institute . 2006 . SAS user’s guide, statistics, version 9.1 . SAS Institute , Cary, NC , 215 pp. Schowalter , T. D . 1995 . Canopy arthropod communities in relation to forest age and alternative harvest practices in western Oregon . Forest Ecol. Manag . 78 : 115 – 125 . Google Scholar CrossRef Search ADS Schowalter , T. D. , and L. M. Ganio . 1998 . Vertical and seasonal variation in canopy arthropod communities in an old-growth conifer forest in southwestern Washington, USA . B. Entomol. Res . 88 : 633 – 640 . Google Scholar CrossRef Search ADS Shaw , E. M. , M. Waddicor , and A. M. Langan . 2006 . Impact of cypermethrin on feeding behaviour and mortality of the spider Pardosa amentata in arenas with artificial ‘vegetation’ . Pest Manag. Sci . 62 : 64 – 68 . Google Scholar CrossRef Search ADS PubMed Souto , D. R. , and K. S. Shields . 2000 . Overview of hemlock health , pp. 76 – 80 . In K. McManus , K. Shields , and D. Souto (eds.), Proceedings: Symposium on sustainable management of hemlock ecosystems in eastern North America , 22–24 June 1999 , Durham, NH . United States Department of Agriculture, Forest Service, Northeastern Research Station Gen. Tech. Rep. NE-267 , Newton Square, PA . Tattar , T. A. , J. A. Dotson , M. S. Ruizzo , and V. B. Steward . 1998 . Translocation of imidacloprid in three tree species when trunk and soil injected . J. Arboric . 24 : 54 – 56 . Ubick , D. , P. Paquin , P. E. Cushing , and V. Roth (eds). 2005 . Spiders of North America: an identification manual . Am. Arach. Soc . 377 , Poughkeepsie, NY. Vickerman , G. P. , and K. D. Sunderland . 1977 . Some effects of dimethoate on arthropods in winter wheat . J. Appl. Ecol . 14 : 767 – 777 . Google Scholar CrossRef Search ADS Wallace , M. S. , and F. P. Hain . 2000 . Field survey and evaluation of native and established predators of the hemlock woolly adelgid (Homoptera: Adelgidae) in the southeastern United States . J. Environ. Entomol . 29 : 638 – 644 . Google Scholar CrossRef Search ADS Ward , J. S. , M. E. Montgomery , C. A. S.-J. Cheah , B. P. Onken , and R. S. Cowles . 2004 . Eastern hemlock forests: guidelines to minimize the impacts of hemlock woolly adelgid . United States Department of Agriculture, Forest Service, Northeastern Area State and Private Forestry, NA-TP-03-04 , Morgantown, WV . Welch , H. , and G. Haddow . 1993 . The world checklist of conifers . Landsman’s Bookshop Ltd ., Herefordshire, UK . Wu , H. , Y. Luo , J. Shi , W. Chen , X. Yan , and P. Jiang . 1989 . Composition and structure of canopy arthropods in Pinus thunbergii . J. Zhejiang For. College . 27 : 501 – 506 . Published by Oxford University Press on behalf of Entomological Society of America 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Environmental Entomology Oxford University Press

Influence of Imidacloprid and Horticultural Oil on Spider Abundance on Eastern Hemlock in the Southern Appalachians

Loading next page...
 
/lp/ou_press/influence-of-imidacloprid-and-horticultural-oil-on-spider-abundance-on-5KRXLmYVFv
Publisher
Oxford University Press
Copyright
Published by Oxford University Press on behalf of Entomological Society of America 2018.
ISSN
0046-225X
eISSN
1938-2936
D.O.I.
10.1093/ee/nvy065
Publisher site
See Article on Publisher Site

Abstract

Abstract Hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae), is an exotic pest of eastern hemlock, Tsuga canadensis (L.) Carrière (Pinales: Pinaceae), in the eastern United States. Two commonly used insecticides to manage adelgid are imidacloprid, a systemic neonicotinoid insecticide, and horticultural oil, a refined petroleum oil foliar spray. We have investigated the influence of imidacloprid and horticultural oil on spider abundance at different canopy strata in eastern hemlock. In total, 2,084 spiders representing 11 families were collected from the canopies of eastern hemlock. In beat-sheet and direct observation samples, the families Theridiidae, Araneidae, Salticidae, and Anyphaenidae were the most abundant. Significantly higher numbers of spiders were recorded on untreated control trees compared with trees treated with imidacloprid using soil drench and soil injection applications. Spider abundance in trees injected with imidacloprid and horticultural oil applications did not significantly differ from control trees. Spider abundance was significantly greater in the top and middle strata of the canopy than in the bottom stratum, where imidacloprid concentrations were the highest. Regression analysis showed that spider abundance was inversely associated with imidacloprid concentration. This research demonstrates that imidacloprid, when applied with selected methods, has the potential to result in reductions of spider densities at different strata. However, slight reductions in spider abundance may be an acceptable short-term ecological impact compared with the loss of an untreated hemlock and all the associated ecological benefits that it provides. Future studies should include investigations of long-term impact of imidacloprid on spiders associated with eastern hemlock. spider, imidacloprid, eastern hemlock, Tsuga canadensis, nontarget Eastern hemlock, Tsuga canadensis (L.) Carrière, is a coniferous tree species native to the eastern United States and adjacent Canadian provinces (Welch and Haddow 1993), where it is an important economic, ecological, and aesthetic component of forest ecosystems (Quimby 1996, Brooks 2001). Hemlocks are long-lived and function as ‘foundation species’, and no other conifer tree species is capable of functionally replacing hemlock in eastern forests (Kizlinski et al. 2002, Orwig et al. 2002, Ellison et al. 2005, Orwig et al. 2008). Populations of eastern hemlock and Carolina hemlock, Tsuga caroliniana Engelmann, have dramatically declined in the eastern United States due to feeding by an invasive insect pest, hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae). All sizes and ages of eastern and Carolina hemlock are susceptible to this adelgid (McClure 1987, Eschtruth et al. 2006). Heavy infestations of hemlock woolly adelgid cause premature foliage drop and bud abortion, and death often occurs within as few as 4 yr after infestation (McClure 1991, McClure et al. 2001, Ward et al. 2004). Besides hemlock woolly adelgid, several other arthropod pests and plant pathogens attack eastern hemlock (Godman and Lancaster 1990, Souto and Shields 2000, Raupp et al. 2004, Ward et al. 2004). These include elongate hemlock scale, Fiorinia externa Ferris (Hemiptera: Diaspididae), hemlock looper, Lambdina fiscellaria fiscellaria (Guenèe) (Lepidoptera: Geometridae), spruce spider mite, Oligonychus ununguis (Jacobi) (Acari: Tetranychidae), hemlock borer, Melanophila fulvogutta (Harris) (Coleoptera: Buprestidae), hemlock rust mite, Nalepella tsugifoliae Keifer (Acari: Eriophyidae), and the pathogen that causes root rot disease, Armillaria mellea (Vahl: Fries) Kummer (Agaricales: Physalacriacaea). Mortality of hemlock in hemlock-dominated forests affects forest structure and composition, as well as ecosystem function. Forest canopies contain highly diversified arthropod species, due to structural heterogeneity and variations in resources (Wu et al. 1989, Schowalter and Ganio 1998). The canopy of eastern hemlock can be expansive and provides suitable habitat for numerous arthropod species. Several studies have investigated the arthropods associated with eastern hemlock, and more than 300 species of insects have been found in hemlock stands in the southern Appalachians (Wallace and Hain 2000, Buck et al. 2005, Lynch et al. 2006, Dilling et al. 2007, Dilling et al. 2009, Jetton et al. 2009). Eastern hemlock also may provide important habitat for arboreal spiders. Spiders are important predators in many temperate forest systems, and their abundance depends mainly upon habitat structure and prey abundance (Halaj et al. 2000). In Oregon, spiders comprised over 50% of the arthropod predators in old growth stands of western hemlock, Tsuga heterophylla (Rafinesque) Sargent (Pinales: Pinaceae) (Schowalter 1995). Additionally, spiders (n = 197 specimens) were among the most abundant arthropod orders collected on eastern hemlocks sampled in the Great Smoky Mountains National Park, comprising 21.7% of all arthropods collected (n = 906 specimens; Falcone and DeWald 2010). In a study comparing spider communities in the canopies of hemlock and deciduous tree species, more spider families and specimens were collected from eastern hemlock (n = 20 families and 3,706 specimens) than from deciduous canopies (n = 15 families and 324 specimens; Mallis and Rieske 2011). These studies, in conjunction with the documentation of insect species that occur in hemlock canopies, demonstrate the diversity of arthropods found in eastern hemlock canopies, and these species may be adversely impacted by loss of hemlock. Several chemical insecticides, including horticultural oil (a refined petroleum oil foliar spray), insecticidal soap, and imidacloprid (a neonicotinoid insecticide), play an important role in suppressing hemlock woolly adelgid and maintaining healthy hemlock trees (Hale 2004, Ward et al. 2004, Dilling et al. 2010, Mayfield et al. 2015, Benton et al. 2016). Although imidacloprid is used widely to manage hemlock woolly adelgid, little is known of the nontarget impacts of this pesticide on predaceous arthropods, especially spiders, in hemlock canopies. Laboratory-based bioassays conducted on two introduced predatory beetle species [Laricobius nigrinus Fender (Coleoptera: Derodontidae) and Sasajiscymnus tsugae (Sasaji and McClure) (Coleoptera: Coccinellidae)] that have been widely released in the eastern United States (Havill et al. 2014) documented lethal effects of imidacloprid, by both ingestion of treated adelgids and direct external contact, as well as sublethal intoxication (Eisenback et al. 2010). Limited information exists on the effect of imidacloprid on introduced or native predators in field settings despite laboratory assays. In one study in the Great Smoky Mountains National Park, no significant differences were observed in arthropod species composition in hemlock stands treated with imidacloprid compared with untreated stands 3–18 mo after treatments (Falcone and DeWald 2010). Also, Mayfield et al. (2015) found no residues of imidacloprid on L. nigrinus collected from imidacloprid-treated trees and no differences in abundance of Laricobius spp. larvae between treated and untreated trees 7 yr following treatment. However, if the abundance of predatory species are reduced on imidacloprid-treated hemlock trees, a greater abundance of some pest species may result. For example, abundance of hemlock rust mite and spruce spider mite was greater on hemlock trees treated with imidacloprid than on untreated trees, which was attributed to the removal of several predators, including spider species, that feed on the mite pests (Raupp et al. 2004). Further investigations on the effect of imidacloprid on predatory species in hemlock canopies are warranted to clarify the impact of imidacloprid, if any, on spider populations. The management of hemlock woolly adelgid using pesticides is warranted, considering the ecological importance of hemlock as a foundation species in forests and its importance as habitat for arthropods. However, lack of information on nontarget effects of these pesticides on predatory arthropods, especially spiders, necessitates additional research to evaluate potential negative impacts. Additionally, imidacloprid movement and persistence throughout foliage in the canopy have been shown to vary greatly throughout different stratum (Dilling et al. 2010, Benton et al. 2015), thus having the potential of varying impacts based on spatial dynamics. The potential impact of these varying levels of imidacloprid on spider abundances in canopies of eastern hemlock is unknown. We have previously reported effects of chemical treatments on nonpredator guilds in the canopy (Dilling et al. 2009, 2010). In this study, we investigated the effect of imidacloprid on vertical distribution of spiders on eastern hemlock. The objective of this study was to determine the effect of insecticide treatments on spider populations at three strata of the canopy of eastern hemlock. Materials and Methods Study Sites This study was conducted at Indian Boundary Recreation Area in the southern part of Cherokee National Forest, Monroe County, Tennessee. To evaluate the vertical stratification of predators in hemlock canopies and determine the impact of insecticides on spiders, on 5 November 2005, eastern hemlocks (n = 30) were selected and arranged into three blocks with 10 trees each (35° 23.787 N, 84° 06. 662 W, elevation: 543 m; 35° 23.764 N, 84° 06.732 W, elevation: 555 m; 35° 24.173 N, 84° 06.268 W, elevation: 565 m). Trees in each block were arranged in five pairs, with one tree in each pair treated with one of five treatments (imidacloprid as tree injection, soil injection, and soil drench; horticultural oil as foliar spray; and untreated control) on one of two application times (29–30 November 2005 [Fall 2005] or 16 April 2006 [Spring 2006]). Tree pairs were assigned based on their similarity in height, condition, and adelgid infestation level. Blocks were located in a short-leaf pine-oak (Type 76) forest (Eyre 1980, Hakeem 2008). Insecticide Treatments The tree injection system consisted of the Mauget 3-ml 10% imicide capsules and feeder tubes (J. J. Mauget Co., Arcadia, CA). Imidacloprid was applied via tree injection at the rate of one capsule per 15 cm dbh (0.15 ml [AI]/2.5 cm dbh). A 1.2-cm hole was made ca. 20.5 cm above ground level using a 0.4-cm bit drilled at a slanting angle. The feeder tube was inserted into the hole, and a capsule was attached to the tube. Capsules were equally distributed on the tree trunk as much as possible for even distribution of insecticide. The capsules were left attached until uptake of the material was complete (ca. 1–5 h). Imidacloprid was applied as a soil injection using a Kioritz soil injector (Kioritz Corp., Tokyo, Japan). Merit 75 WP insecticide (Bayer, Kansas City, MO) was diluted to 1 g (AI)/2.5 cm dbh in 60 ml of water. Soil injection was applied within 45 cm of the base of the trunk and spaced as evenly as possible. The insecticide was injected at a depth of 7 cm below soil surface at 30 ml per injection. The soil drench treatment, applied using a high-pressure hydraulic sprayer (FMC Corporation, Jonesboro, AR), consisted of imidacloprid (Merit 75 WP) at the rate of 1.5 g (AI)/2.5 cm dbh. The standard recommended dose of 50 g of imidacloprid was mixed with 379-liter water. The soil beneath the drip line of the canopy was sprayed with 125 liters of mixture to provide maximum coverage. SunSpray horticultural oil (Sun Co., Philadelphia, PA) was applied as a foliar spray using a high-pressure hydraulic sprayer (FMC Corporation). The horticultural oil foliar spray was applied at the rate of 7.6 liters (AI)/379 liters of water and sprayed on the canopy until runoff to ensure that the tree was fully covered. Sampling of Vertical Strata Sampling of spiders in the canopy of eastern hemlock was conducted twice: 16–19 August 2006 and 5–11 September 2007. Each tree canopy was divided into three equal vertical sections, or strata, by height: top (upper 1/3 of canopy), middle (middle 1/3 of canopy), and bottom (lower 1/3 of canopy). A bucket truck (Genie Z 45/22, Tigard, OR) was used to collect samples from the three strata of each tree. Beat-sheet sampling and handpicking were the collection methods used to collect spiders from the canopy. From each stratum, three random beat-sheet samples (0.7 × 0.7 m; Bioquip Products, Rancho Dominguez, CA) were taken by holding the beat sheet underneath the branch and striking the branch five times with a wooden rod (1 m). Handpicking specimens involved collecting visually observed specimens for 5 min per stratum using forceps or fingers. Specimens collected during both sampling methods were placed in vials containing 75% alcohol and taken to the laboratory for processing and identification. Spiders were identified, where possible, by the author (AH) to family using Howel and Jenkins (2004) and Ubick et al. (2005), and family identifications were confirmed by Pat Barnwell, University of Tennessee. The specimens are housed in the Integrated Pest Management and Biological Control Laboratory, the University of Tennessee, Knoxville. Quantification of Imidacloprid Imidacloprid concentrations in twig and needle tissue were determined in a synchronous study by Dilling et al. (2010). Commercially available enzyme-linked immunosorbent assay kits were used to measure the imidacloprid residues within branches to determine the potential effect of imidacloprid on spiders (EnviroLogix 2015). In this test, the compound horseradish peroxidase-labeled imidacloprid was used, which competes with the imidacloprid residues present in the sample for a limited number of antibody sites on the walls of the test wells. This kit was used to quantify concentrations of imidacloprid between 0.2 and 6.0 ppb. Sample preparation and analysis are detailed in Dilling et al. (2010). Data Analysis Data were analyzed as a randomized block design with two fixed factors (treatment and application time) and two repeated measures (sampling year and strata). Specimens collected from beat-sheet sampling and handpicking were pooled to represent spider abundance per stratum per tree per sampling year. Treatment, application time, stratum, and sampling year were considered independent variables, and spider abundance per stratum per tree per sampling year was used as the response variable. All interactions among these variables were analyzed for significant associations. Shapiro–Wilk W test for normality and Levene’s test of homogeneity of variance were used to verify that data conformed to the assumptions of analysis of variance (ANOVA). Due to heterogeneity of group variances, data were rank transformed, and all data were analyzed by one-way ANOVA (ANOVA; PROC MIXED). Significance for all ANOVA tests was set at P ≤ 0.05. When significant ANOVA results were obtained, post hoc analysis to detect differences among means was determined using Tukey–Kramer test (α = 0.05). Kenward-–Roger method was used to adjust degrees of freedom for repeated measures because repeated measurements are correlated and consequently require an adjustment in the error degrees of freedom used in calculating significance tests. Original means and standard error values are reported. Similarity in family diversity and abundance among samples representing community composition was determined using analysis of similarity (ANOSIM) on Bray–Curtis distances constructed in a resemblance matrix using PRIMER version 6 software (Clarke and Warwick 1997) across tree strata and treatment type. Differences in samples were plotted using nonmetric multidimensional scaling. The relationship between imidacloprid concentrations at each stratum and mean spider specimen abundance was analyzed with linear regression analysis using PROC REG. Mean imidacloprid concentrations in parts per billion (ppb) determined from composite twig and branch samples presented in Dilling et al. (2010) that coincide with the vertical sampling dates presented in this study were regressed with the respective mean spider abundance for each stratum, treatment, and application time. All data analyses were conducted using SAS/STAT software, version 9.1 of the SAS System for Windows (SAS Institute 2006). Results and Discussion Density and Diversity of Spiders A diverse group of spiders was recovered from canopies of eastern hemlock. In total, 2,091 spiders were collected from canopies of eastern hemlock during this study. Of these, 2,084 spiders belong to 11 families; the families of 7 individuals could not be determined (Table 1). In general, spiders were distributed throughout the tree strata. Higher numbers (n = 1,588) of spiders were collected in beat-sheet samples compared with direct observations (n = 503), and greater numbers of spiders were collected in 2006 (n = 1,144) than in 2007 (n = 947). The most abundant families were Theridiidae, Araneidae, Salticidae, and Anyphaenidae. These four families represented 83.5% of all spiders collected (Table 1). Table 1. Spider families recorded from different strata on eastern hemlock, Indian Boundary Campground, Cherokee National Forest, Tennessee, 2006 and 2007 Spider family Stratum* Treatment† Total abundance‡ 2006 2007 2006 2007 Araneidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 445 Anyphaenidae B, M, T B CO, HO, SD, SI, TI CO, HO, SD, SI, TI 319 Corinnidae M SI, TI 2 Clubionidae B, T CO, SD, SI, TI HO, TI 16 Dictynidae B, T B, M CO, SD, SI, TI TI, SI 6 Linyphiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Salticidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 374 Tetragnathidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Theridiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 609 Thomisidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 186 Uloboridae B, T B, M CO, SD, SI, TI HO, TI 13 Unknown B, T B, T CO, HO, SD, SI, TI CO, TI 7 Spider family Stratum* Treatment† Total abundance‡ 2006 2007 2006 2007 Araneidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 445 Anyphaenidae B, M, T B CO, HO, SD, SI, TI CO, HO, SD, SI, TI 319 Corinnidae M SI, TI 2 Clubionidae B, T CO, SD, SI, TI HO, TI 16 Dictynidae B, T B, M CO, SD, SI, TI TI, SI 6 Linyphiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Salticidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 374 Tetragnathidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Theridiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 609 Thomisidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 186 Uloboridae B, T B, M CO, SD, SI, TI HO, TI 13 Unknown B, T B, T CO, HO, SD, SI, TI CO, TI 7 * Strata from which spider families were collected: B = bottom; M = middle; T = top. † Treatments from which spider families were collected: CO = control; HO = horticultural oil; SD = soil drench; SI = soil injection; TI = tree injection. ‡ Total abundance represents the number of all individuals collected in each family. View Large Table 1. Spider families recorded from different strata on eastern hemlock, Indian Boundary Campground, Cherokee National Forest, Tennessee, 2006 and 2007 Spider family Stratum* Treatment† Total abundance‡ 2006 2007 2006 2007 Araneidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 445 Anyphaenidae B, M, T B CO, HO, SD, SI, TI CO, HO, SD, SI, TI 319 Corinnidae M SI, TI 2 Clubionidae B, T CO, SD, SI, TI HO, TI 16 Dictynidae B, T B, M CO, SD, SI, TI TI, SI 6 Linyphiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Salticidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 374 Tetragnathidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Theridiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 609 Thomisidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 186 Uloboridae B, T B, M CO, SD, SI, TI HO, TI 13 Unknown B, T B, T CO, HO, SD, SI, TI CO, TI 7 Spider family Stratum* Treatment† Total abundance‡ 2006 2007 2006 2007 Araneidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 445 Anyphaenidae B, M, T B CO, HO, SD, SI, TI CO, HO, SD, SI, TI 319 Corinnidae M SI, TI 2 Clubionidae B, T CO, SD, SI, TI HO, TI 16 Dictynidae B, T B, M CO, SD, SI, TI TI, SI 6 Linyphiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Salticidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 374 Tetragnathidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, TI 57 Theridiidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 609 Thomisidae B, M, T B, M, T CO, HO, SD, SI, TI CO, HO, SD, SI, TI 186 Uloboridae B, T B, M CO, SD, SI, TI HO, TI 13 Unknown B, T B, T CO, HO, SD, SI, TI CO, TI 7 * Strata from which spider families were collected: B = bottom; M = middle; T = top. † Treatments from which spider families were collected: CO = control; HO = horticultural oil; SD = soil drench; SI = soil injection; TI = tree injection. ‡ Total abundance represents the number of all individuals collected in each family. View Large The number of specimens and spider families in this study are similar to those observed in other studies of spiders associated with eastern hemlock. In mature stands of eastern hemlock in the Great Smoky Mountains National Park, more spider families (n = 16) but fewer specimens (n = 191) were collected from pitfall traps (Jetton et al. 2009). However, limited direct comparison can be made between this study and the current study, as spider families and specimen abundance in terrestrial systems could vary greatly from what is present in canopies. Greater spider specimen abundance was observed in the current study compared with that observed in hemlock canopies in another study in the Great Smoky Mountains National Park (n = 197; Falcone and DeWald 2010). However, family identifications were not performed for spiders collected in hemlock canopies in the Smokies (Falcone and DeWald 2010), and family numbers cannot be compared to this study. In a study comparing spider populations of hemlock canopies versus deciduous canopies in Kentucky, more specimens (n = 3,706) and families (n = 20) were collected from untreated hemlock canopies than in the current study (Mallis and Rieske 2011). The number of samples (24 vs 2 samples) may have been the greatest influence on the differences in spiders observed in Kentucky compared to this study. Sampling was conducted monthly over a 24-mo period in Kentucky (Mallis and Rieske 2011), whereas intensive spider sampling was conducted two times during this study period. The hemlock canopies evaluated as part of the current study contained an abundance of prey items (293 insect species; Dilling et al. 2009), with similar orders (Diptera, Psocoptera, Hymenoptera, and Hemiptera) comprising the bulk of prey items observed in Kentucky (Mallis and Rieske 2010). Therefore, while more spider families and specimens were observed by Mallis and Rieske (2011), family numbers and specimen abundance similar to those observed in Kentucky are expected from these study trees if sampling were conducted at similar frequencies. Vertical Distribution of Spider Families on Eastern Hemlock Community composition of families was not significantly different among tree strata (R = 0.054, n = 30, P > 0.001 or treatment types (R = 0.134, n = 30, P > 0.001). The similarity of family community composition across bottom, middle, and top strata indicates a homogenous distribution throughout the tree canopy. The similarity of family community composition across treatment types indicates a negligible impact on the community composition of spiders associated with eastern hemlock. This homogenous distribution could be due to vertical distribution of prey. Although community composition of family was similar among strata, the abundance of individuals within families, which will be discussed later, was distributed differently among strata. Impact of Insecticidal Treatments on Spiders Insecticidal treatments influenced the numbers of spiders collected from eastern hemlock. Significant differences (F4,19.9 = 5.73; P < 0.0031) in mean spider abundance per treatment per tree were observed among control (no treatments), soil drench, and soil injection (Table 2). In both sampling years, mean spider abundance in soil drench and soil injection treatments was significantly lower than control treatments, but no differences in spider abundance were observed among control trees, tree injection, and horticultural oil treatments (Table 2). Specimen abundance in samples collected in 2006 ranged from 19.88 spiders per tree in control treatments to 7.17 spiders per tree in soil drench treatments. Specimen abundance in samples collected in 2007 ranged from 17.41 spiders per tree in control treatments to 6.76 spiders per tree in soil injection treatments. Spider abundance was numerically greater for all treatments in 2006 than in 2007, with the exception of soil drench treatments, which were numerically greater in 2007 than in 2006. Spider abundance in soil injection and soil drench treatments also was significantly lower than control treatments for the overall mean of both years combined, with overall spider abundance per tree ranging from 18.64 in control treatments to 3.79 in soil injection treatments. However, no differences in spider abundance were observed among control trees, tree injection, and horticultural oil treatments for either sampling year or the overall mean of both years (Table 2). Additionally, in both sampling years and for the overall mean of both years combined, no significant differences were observed within imidacloprid treatments (tree injection, soil injection, and soil drench). No significant interactions were observed among treatment, application time, sampling year, stratum, or all. Table 2. Comparison of spider abundance (mean ± SE) on eastern hemlock, Tsuga canadensis, treated with different insecticides, 2006 and 2007 Treatment Year Average per year 2006* 2007 Control (not treated) 19.88 ± 2.32aA 17.41 ± 1.74aA 18.64 ± 2.03aA Tree injection 15.33 ± 1.85abA 13.83 ± 2.58abA 14.58 ± 2.21abA Horticultural oil 13.44 ± 2.57abA 10.63 ± 1.73abA 12.03 ± 2.15abA Soil drench 7.17 ± 0.78bA 7.80 ± 1.58bA 7.48 ± 1.18bA Soil injection 8.83 ± 1.01bA 6.76 ± 1.02bA 3.79 ± 1.01bA Treatment Year Average per year 2006* 2007 Control (not treated) 19.88 ± 2.32aA 17.41 ± 1.74aA 18.64 ± 2.03aA Tree injection 15.33 ± 1.85abA 13.83 ± 2.58abA 14.58 ± 2.21abA Horticultural oil 13.44 ± 2.57abA 10.63 ± 1.73abA 12.03 ± 2.15abA Soil drench 7.17 ± 0.78bA 7.80 ± 1.58bA 7.48 ± 1.18bA Soil injection 8.83 ± 1.01bA 6.76 ± 1.02bA 3.79 ± 1.01bA *Means followed by the same lowercase letter(s) within the same column, and means followed by the same uppercase letter within the same row are not significantly different (P ≤ 0.05; Tukey–Kramer test). View Large Table 2. Comparison of spider abundance (mean ± SE) on eastern hemlock, Tsuga canadensis, treated with different insecticides, 2006 and 2007 Treatment Year Average per year 2006* 2007 Control (not treated) 19.88 ± 2.32aA 17.41 ± 1.74aA 18.64 ± 2.03aA Tree injection 15.33 ± 1.85abA 13.83 ± 2.58abA 14.58 ± 2.21abA Horticultural oil 13.44 ± 2.57abA 10.63 ± 1.73abA 12.03 ± 2.15abA Soil drench 7.17 ± 0.78bA 7.80 ± 1.58bA 7.48 ± 1.18bA Soil injection 8.83 ± 1.01bA 6.76 ± 1.02bA 3.79 ± 1.01bA Treatment Year Average per year 2006* 2007 Control (not treated) 19.88 ± 2.32aA 17.41 ± 1.74aA 18.64 ± 2.03aA Tree injection 15.33 ± 1.85abA 13.83 ± 2.58abA 14.58 ± 2.21abA Horticultural oil 13.44 ± 2.57abA 10.63 ± 1.73abA 12.03 ± 2.15abA Soil drench 7.17 ± 0.78bA 7.80 ± 1.58bA 7.48 ± 1.18bA Soil injection 8.83 ± 1.01bA 6.76 ± 1.02bA 3.79 ± 1.01bA *Means followed by the same lowercase letter(s) within the same column, and means followed by the same uppercase letter within the same row are not significantly different (P ≤ 0.05; Tukey–Kramer test). View Large These results indicate that selected methods of imidacloprid applications can cause a short-term reduction in spider populations. Furthermore, no differences observed between sampling years for all treatments, in conjunction with no significant interactions between application time and treatment, suggests that timing of applications has no short-term influence on spider abundance. During both sampling years, spiders were least abundant in trees treated with imidacloprid applied with a soil injection or soil drench compared with control, horticultural oil, and trunk injection treatments. These reductions in spider numbers are the first documentation of short-term, nontarget impacts of imidacloprid on spiders in eastern hemlock canopies. Although the long-term impacts of imidacloprid on spider populations are unknown, these populations are expected to rebound over time as imidacloprid concentrations lessen in the plant tissues. These results differ from those reported spider populations in the Great Smoky Mountains National Park, which showed no differences in spider densities between treated and untreated hemlock trees (Falcone and DeWald 2010). In their study, they applied imidacloprid as a soil drench, while the current study had three imidacloprid application types and a foliar application of horticultural oil. The interval between application and sampling was similar in both studies, with the sampling interval ranging from ca. 1 mo (at one of six treated sites) to 18 mo in the Great Smoky Mountain National Park and from ca. 4 mo to ca. 22 mo in the current study. However, the total number of samples (n = 240) taken during the Great Smoky Mountains National Park study was less than the total number of samples in the current study (n = 720). In their study, spiders were sampled by branch clipping instead of beat-sheet sampling or handpicking, and branch clipping may not be an adequately intensive sampling method to properly assess spider abundance in forest settings, as it may allow some spiders the opportunity to flee during sampling. The combination of fewer samples and branch clipping as a sampling technique may have influenced the numbers of spiders collected in the Great Smoky Mountains National Park and prevent adequate comparisons to our findings. Trees treated with imidacloprid as a trunk injection consistently had greater numbers of spiders than horticultural oil, soil injection, or soil drench treatments. Trunk injection requires the trunk of trees to be drilled to allow insertion of the imidacloprid delivery apparatus. However, the lack of reductions of spiders on trees treated with imidacloprid applied as a trunk injection is unclear. Although no significant differences (F8,92.6 = 1.99; P = 0.0556) were observed among interactions of treatment, application time, and stratum, the trend of soil injection and soil drench treatments containing fewer spiders than other treatments is evident (Fig. 1). In fall and spring application times and all strata, soil injection and soil drench treatments were the lowest, with the exception of trees treated with horticultural oil in Spring 2006. For these trees, spider abundance was lower in horticultural oil treatments than soil injection, as well as tree injection and control treatments. The reason for this low level of spider abundance in trees treated with horticultural oil at one stratum for one application time is unclear. These nonsignificant results may be due to the complex nature of interactions between treatments, season, and strata, although treatments and strata were significant (Fig. 1). Fig. 1. View largeDownload slide Comparison of spider abundance (mean ± SE) within each treatment and strata for hemlock trees treated in Fall 2005 and Spring 2006. No significant differences (F8,92.6 = 1.99; P = 0.0556) were observed in interactions among treatment, application time, strata, or all. Fig. 1. View largeDownload slide Comparison of spider abundance (mean ± SE) within each treatment and strata for hemlock trees treated in Fall 2005 and Spring 2006. No significant differences (F8,92.6 = 1.99; P = 0.0556) were observed in interactions among treatment, application time, strata, or all. Canopy location (i.e., stratum) also influenced spider abundance. Significant differences (F2,92.6 = 5.60; P < 0.0051) were observed between mean spider abundance per sample per year between the top and the middle strata compared with the bottom stratum (Table 3). In both sampling years, significantly lower numbers of spiders were observed from bottom stratum than either the top or middle stratum, but no differences were observed in spider abundance in each respective stratum between years (Table 3). Specimen abundance in samples collected in 2006 ranged from 14.00 spiders per tree in the top stratum to 11.07 spiders per tree in the bottom stratum. Specimen abundance in samples collected in 2007 ranged from 11.56 spiders per tree in the top stratum to 8.56 spiders per tree in the bottom stratum. Significantly lower numbers of spiders also were recovered from the bottom stratum than from the top or middle stratum for the overall mean of both years combined, with spider abundance per tree ranging from 12.78 in the top stratum to 9.81 in the bottom stratum (Table 3). Higher numbers of spiders in the top strata is not surprising, as Joseph et al. (2011) also reported higher numbers of hemlock woolly adelgid ovisacs in the upper tree crown, suggesting less imidacloprid concentrations. No significant interactions were observed among stratum, treatment, application time, sampling year, or all. Oguri et al. (2014) compared spider assemblage between the upper and lower canopy layers (i.e., strata), and between the canopy and forest floor in evergreen cedar, Cryptomeria japonica D. Don (Pinales: Cupressaceae), and deciduous larch, Larix kaempferi [Lamb.] Carrieré (Pinales: Pinaceae), and found significant differences in number of species between strata in L. kaempferi only. Table 3. Comparison of spider abundance (mean ± SE) from three strata of eastern hemlock, Tsuga canadensis, 2006 and 2007 Strata Year Average per year 2006* 2007 Top 14.00 ± 1.95aA 11.56 ± 1.26aA 12.78 ± 1.60aA Middle 13.52 ± 1.57aA 13.72 ± 1.88aA 13.62 ± 1.72aA Bottom 11.07 ± 1.27bA 8.56 ± 1.40bA 9.81 ± 1.33bA Strata Year Average per year 2006* 2007 Top 14.00 ± 1.95aA 11.56 ± 1.26aA 12.78 ± 1.60aA Middle 13.52 ± 1.57aA 13.72 ± 1.88aA 13.62 ± 1.72aA Bottom 11.07 ± 1.27bA 8.56 ± 1.40bA 9.81 ± 1.33bA *Means followed by the same lowercase letter within the same column, and means followed by the same uppercase letter within the same row are not significantly different (P ≤ 0.05; Tukey–Kramer test). View Large Table 3. Comparison of spider abundance (mean ± SE) from three strata of eastern hemlock, Tsuga canadensis, 2006 and 2007 Strata Year Average per year 2006* 2007 Top 14.00 ± 1.95aA 11.56 ± 1.26aA 12.78 ± 1.60aA Middle 13.52 ± 1.57aA 13.72 ± 1.88aA 13.62 ± 1.72aA Bottom 11.07 ± 1.27bA 8.56 ± 1.40bA 9.81 ± 1.33bA Strata Year Average per year 2006* 2007 Top 14.00 ± 1.95aA 11.56 ± 1.26aA 12.78 ± 1.60aA Middle 13.52 ± 1.57aA 13.72 ± 1.88aA 13.62 ± 1.72aA Bottom 11.07 ± 1.27bA 8.56 ± 1.40bA 9.81 ± 1.33bA *Means followed by the same lowercase letter within the same column, and means followed by the same uppercase letter within the same row are not significantly different (P ≤ 0.05; Tukey–Kramer test). View Large Impact of Imidacloprid Concentrations on Spider Abundance Spider abundance in the canopy strata was directly associated with imidacloprid concentrations, as linear regression showed a strong negative association between mean specimen abundance across both application times and treatments and mean imidacloprid concentrations across all strata in both sampling years (Figs. 2 and 3). Different application methods of imidacloprid resulted in significant differences in imidacloprid concentrations distributed throughout the canopy, but for all imidacloprid application methods concentrations were highest in the bottom stratum and decreased in the middle and top strata (Dilling et al. 2010). In the top stratum, imidacloprid concentrations ranged from 0 to 98.44 ppb across all sampling times and treatments (Figs. 2A and D, 3A and D). Mean specimen abundance in the top stratum ranged from 1.62 to 5.19 per application time per sampling year per treatment per tree across all sampling times and treatments and declined as imidacloprid concentrations increased (Figs. 2A and D, 3A and D). In the middle stratum, imidacloprid concentrations ranged from 0 to 162.98 ppb across all sampling times (Figs. 2B and E, 3B and E). Mean specimen abundance in the middle stratum ranged from 0.48 to 5.15 per application time per sampling year per treatment per tree across all sampling times and treatments (Figs. 2B and E, 3B and E). Imidacloprid concentrations in the bottom stratum ranged from 0 to 199.33 ppb across all sampling times (Figs. 1C and F,2C and F). Mean specimen abundance in the bottom stratum ranged from 0.23 to 5.42 per application time per sampling year per treatment per tree across all sampling times and application types (Figs. 2C and F, 3C and F). Fig. 2. View largeDownload slide Relationship between mean spider abundance per tree based on the average number of specimens collected from trees treated in Fall 2005 within the top (A), middle (B), and bottom (C) strata, trees treated in Spring 2006 within the top (D), middle (E), and bottom (F) strata, and the mean imidacloprid concentrations (ppb) within each respective stratum for samples collected in August 2006. Fig. 2. View largeDownload slide Relationship between mean spider abundance per tree based on the average number of specimens collected from trees treated in Fall 2005 within the top (A), middle (B), and bottom (C) strata, trees treated in Spring 2006 within the top (D), middle (E), and bottom (F) strata, and the mean imidacloprid concentrations (ppb) within each respective stratum for samples collected in August 2006. Fig. 3. View largeDownload slide Relationship between mean spider abundance per tree based on the average number of specimens collected from trees treated in Fall 2005 within the top (A), middle (B), and bottom (C) strata, trees treated in Spring 2006 within the top (D), middle (E), and bottom (F) strata, and the mean imidacloprid concentrations (ppb) within each respective stratum for samples collected in September 2007. Fig. 3. View largeDownload slide Relationship between mean spider abundance per tree based on the average number of specimens collected from trees treated in Fall 2005 within the top (A), middle (B), and bottom (C) strata, trees treated in Spring 2006 within the top (D), middle (E), and bottom (F) strata, and the mean imidacloprid concentrations (ppb) within each respective stratum for samples collected in September 2007. These results indicate that the stratum in which spiders occur significantly influenced their abundance. Also, no differences observed between sampling years for all strata, in conjunction with no significant interactions between application time and strata, suggest that timing of applications do not influence short-term spider abundance in strata. During both sampling years and for the mean of both years combined, spiders were least abundant in the bottom stratum. Low numbers of spiders on the bottom stratum were directly associated with higher concentrations of imidacloprid in the bottom stratum, and high R2 values for all regressions demonstrate the strong associations between spider abundance within treatments and concentrations of imidacloprid at different strata. Therefore, certain methods of imidacloprid application (soil injection and soil drench) may not only reduce spider densities, but can have varying impacts on spiders throughout the canopy of eastern hemlock. The association between spider abundance and concentrations of imidacloprid observed during this study provides insight into what concentrations may influence spider abundance in the field. Imidacloprid concentrations of 120–250 ppb have been reported to suppress populations of hemlock woolly adelgid over time and cause lethal and sublethal effects to the adelgid predators L. nigrinus and S. tsugae (Tattar et al. 1998, Cowles et al. 2006, Eisenback et al. 2010). However, the accumulation of imidacloprid in prey items other than hemlock woolly adelgid that may cause lethal and sublethal effects in spiders is unknown. Regression analyses in the current study suggest that imidacloprid concentrations in twig and needle tissues >120 ppb may predict decreases in spider populations. For example, spider abundance on soil drenched trees treated in Spring 2006 and sampled in August 2006 (Fig. 2) was 2.20 spiders per tree at a concentration of 96.25 ppb in the top strata, yet only 0.62 spiders per tree at a concentration of 159.33 ppb in the middle strata. Furthermore, imidacloprid concentrations in all samples from the bottom stratum of soil injection- and soil drench-treated trees exceeded 120 ppb, with the lowest concentration (131.29 ppb) observed in soil injection-treated trees treated in Spring 2006 and sampled in September 2007 (Fig. 3). Imidacloprid concentrations do not exceed ca. 100 ppb in the top strata of the tree, and spider populations occurring in this stratum may not have experienced reductions during the study following imidacloprid treatments. Imidacloprid can translocate into the canopy quickly and have sustained effect on herbivores. Dilling et al. (2010) observed peaks in imidacloprid concentrations 9–12 mo following application, yet 3 mo following soil injection and soil drench applications concentrations in the bottom stratum of hemlocks were >169 ppb. Additionally, the bottom stratum maintained concentrations of imidacloprid >125 ppb 24 mo following soil injection and soil drench applications. Because imidacloprid can reach concentrations >120 ppb relatively quickly and sustain those concentrations in plant tissues over time, nontarget impacts to predatory species within the first 2 yr following imidacloprid application may be unavoidable. Nontarget impacts of imidacloprid on spider abundance are not surprising, as several studies have documented reduced spider abundance caused by applications of chemical pesticides in agricultural settings (Vickerman and Sunderland 1977, Salem and Matter 1991, Pekar 1999, Epstein et al. 2000, Meissle and Lang 2005, Cardenas et al. 2006, Markó et al. 2009). Furthermore, the presence and concentration of imidacloprid could influence spider numbers in several ways. First, direct contact with insecticides may be toxic to spiders and cause a reduction in spider abundance (Shaw et al. 2006). However, in the current study, due to the application methods used, it is unlikely that spiders had direct contact with imidacloprid. Second, imidacloprid could decrease numbers of herbivorous arthropod prey species, causing spiders to starve or leave the area in search of other prey. Dilling et al. (2009) observed significant reductions in specimen abundance in several insect guilds, including phytophagous insects, in trees where imidacloprid was applied as soil drench. These observed reductions could have increased competition among spiders, causing some to starve and/or cuing some spiders to leave in search of prey. Third, spiders may acquire imidacloprid in prey that have recently consumed treated plant material, causing secondary poisoning (Epstein et al. 2000). This secondary poisoning could lead to sublethal intoxication or death, depending on the concentration of the pesticide in the prey. Sublethal intoxication can cause reductions in adult longevity, fecundity, fertilization, and inhibition of feeding that can lead to decreases in specimen abundance (Devine et al. 1996, Boina et al. 2009). Decreases in spider abundance observed in the current study are most likely due to sublethal and/or lethal intoxication; however, more research is needed to determine the tritrophic movement of imidacloprid in this system. Spiders are generalist predators and feed on several pest species. Nontarget impact of pesticides that influence their abundance on hemlock may lead to further changes in arthropod communities associated with eastern hemlock. When considering applications of imidacloprid, dose, concentration, and application method should be considered to reduce nontarget impact on spiders. The findings reported herein demonstrate that imidacloprid applied as a soil injection or soil drench at the recommended rates can cause short-term reductions in spider populations in the lower canopy where imidacloprid concentrations are highest. Although the long-term effects of imidacloprid on spiders and other predatory arthropods are unknown, millions of untreated eastern hemlocks have died or are severely damaged, drastically altering hemlock and associated forest ecosystems. Imidacloprid provides an effective, economical treatment to protect and preserve some trees. Thus, slight reductions in spider abundance may be an acceptable short-term ecological impact compared with the loss of an untreated hemlock and all the associated ecological benefits that it provides. Monitoring of spider populations on these study trees over time, in conjunction with quantification of imidacloprid concentrations at the time of sampling, would document the persistence of imidacloprid in treated trees, as well as assess the impact of imidacloprid on spider population years after treatment. This information could inform conservation-oriented management strategies that are aimed at maintaining healthy canopies by reducing the impact of hemlock woolly adelgid, while simultaneously preserving diversity of predatory arthropods within the canopy of eastern hemlock. Acknowledgments We extend our special thanks to Joshua Grant, David Paulsen, Nicholas Reynolds, and Renee Follum (Department of Entomology and Plant Pathology, University of Tennessee, Knoxville, TN) for their assistance in collection of samples, Pat Barnwell for confirming family identifications, Ann Reed (Statistical Consulting Center, University of Tennessee, Knoxville, TN) for statistical assistance, and personnel at the Tellico Plains Ranger Station (Cherokee National Forest) for allowing us to conduct this study at Indian Boundary. This study was funded in part by USDA Forest Service (FS agreement no. 05-PA-11083150-070). References Cited Benton , E. P. , J. F. Grant , R. J. Webster , R. J. Nichols , R. S. Cowles , A. F. Lagalante , and C. I. Coots . 2015 . Assessment of imidacloprid and its metabolites in foliage of eastern hemlock multiple years following treatment for hemlock woolly adelgid, Adelges tsugae (Hemiptera: Adelgidae), in forested conditions . J. Econ. Entomol . 108 : 2672 – 2682 . Google Scholar CrossRef Search ADS PubMed Benton , E. P. , J. Grant , R. Cowles , J. Webster , R. Nichols , A. Lagalante , and C. Coots . 2016 . Assessing relationships between tree diameter and long-term persistence of imidacloprid and olefin to optimize imidacloprid treatments on eastern hemlock . Forest Ecol. Manag . 370 : 12 – 21 . Google Scholar CrossRef Search ADS Boina , D. R. , E. O. Onagbola , M. Salyani , and L. L. Stelinski . 2009 . Antifeedant and sublethal effects of imidacloprid on Asian citrus psyllid, Diaphorina citri . Pest Manag. Sci . 65 : 870 – 877 . Google Scholar CrossRef Search ADS PubMed Brooks , R. T . 2001 . Effects of the removal of overstory hemlock from hemlock-dominated forest on eastern redback salamanders . Forest Ecol. Manag . 149 : 197 – 204 . Google Scholar CrossRef Search ADS Buck , L. , P. Lambdin , D. Paulsen , J. Grant , and A. Saxton . 2005 . Checklist of insect species associated with eastern hemlock in the Great Smoky Mountains National Park and environs . Tenn. Acad. Sci . 80 : 1 – 10 . Cardenas , M. , F. Ruano , P. Garcia , F. Pascual , and M. Campos . 2006 . Impact of agricultural management on spider populations in the canopy of olive trees . Biol. Control 38 : 188 – 195 . Google Scholar CrossRef Search ADS Clarke , K. R. , and R. M. Warwick . 1997 . PRIMER user manual (Plymouth routines in multivariate ecological research) . Plymouth Marine Laboratory , Plymouth, UK . Cowles , R. S. , M. E. Montgomery , and C. A. S.-J. Cheah . 2006 . Activity and residues of imidacloprid applied to soil and tree trunks to control hemlock woolly adelgid (Hemiptera: Adelgidae) in forests . J. Econ. Entomol . 99 : 1259 – 1267 . Google Scholar CrossRef Search ADS Devine , G. J. , Z. K. Harlings , A. W. Scarr , and A. L. Devonshire . 1996 . Lethal and sublethal effects of imidacloprid on nicotine-tolerant Myzus nicotianae and Myzus persicae . Pestic. Sci . 48 : 57 – 62 . Google Scholar CrossRef Search ADS Dilling , C. , P. Lambdin , J. Grant , and L. Buck . 2007 . Insect guild structure associated with eastern hemlock in the southern Appalachians . Environ. Entomol . 36 : 1408 – 1414 . Google Scholar CrossRef Search ADS PubMed Dilling , C. , P. Lambdin , J. Grant , and R. Rhea . 2009 . Community response of insects associated with eastern hemlock to imidacloprid and horticultural oil treatments . Environ. Entomol . 38 : 53 – 66 . Google Scholar CrossRef Search ADS PubMed Dilling , C. , P. Lambdin , J. Grant , and R. Rhea . 2010 . Spatial and temporal distribution of imidacloprid in eastern hemlock in the southern Appalachians . J. Econ. Entomol . 103 : 368 – 373 . Google Scholar CrossRef Search ADS PubMed Eisenback , B. M. , S. M. Salom , L. T. Kok , and A. F. Lagalante . 2010 . Lethal and sublethal effects of imidacloprid on hemlock woolly adelgid (Hemiptera: Adelgidae) and two introduced predator species . J. Econ. Entomol . 103 : 1222 – 1234 . Google Scholar CrossRef Search ADS PubMed Ellison , A. M. , M. S. Bank , D. C. Barton , E. A. Colburn , K. Elliott , C. R. Ford , D. R. Foster , B. D. Kloeppel , J. D. Knoepp , G. M. Lovett , et al. 2005 . Loss of foundation species: consequences for the structure and dynamics of forested ecosystems . Front. Ecol. Environ . 3 : 479 – 486 . Google Scholar CrossRef Search ADS EnviroLogix . 2015 . Quantitative laboratory test for the detection of imidacloprid in water. Technical data sheet for Microwell Plate Assay, cat. No. EP006 . Retrieved April 25, 2018 from https://www.envirologix.com/wp-content/uploads/2015/05/EP006-Imidacloprid-101215.pdf. Epstein , D. L. , R. S. Zack , J. F. Brunner , L. Gut , and J. J. Brown . 2000 . Effects of broad-spectrum insecticides on epigeal arthropod biodiversity in Pacific Northwest apple orchards . Environ. Entomol . 29 : 340 – 348 . Google Scholar CrossRef Search ADS Eschtruth , A. K. , N. L. Cleavitt , J. J. Battles , R. A. Evans , and T. J. Fahey . 2006 . Vegetation dynamics in declining eastern hemlock stands: 9 years of forest response to hemlock woolly adelgid infestation . Can. J. Forest Res . 36 : 1435 – 1450 . Google Scholar CrossRef Search ADS Eyre , F . 1980 . Forest cover types of the United States and Canada . Society of American Foresters , Washington, DC . Falcone , J. F. , and L. E. DeWald . 2010 . Comparisons of arthropod and avian assemblages in insecticide-treated and untreated eastern hemlock (Tsuga canadensis [L.] Carr.) stands in Great Smoky Mountains National Park, USA . Forest Ecol. Manag . 260 : 856 – 863 . Google Scholar CrossRef Search ADS Godman , R. M. , and K. Lancaster . 1990 . Tsuga canadensis . In R. M. Burns and B. H. Honkala (eds.), Silvics of North America , vol. 1 . United States Department of Agriculture, Forest Service Hdbk. 654 , Washington, DC . Hakeem , A . 2008 . Non-target effect of imidacloprid on predatory guild on eastern hemlock, Tsuga canadensis (L.) Carriere, in the southern Appalachians . M.S. thesis, University of Tennessee , Knoxville , 78 pp. Halaj , J. , D. W. Ross , and A. R. Moldenke . 2000 . Importance of habitat structure to the arthropod food-web in douglas-fir canopies . Oikos . 90 : 139 – 152 . Google Scholar CrossRef Search ADS Hale , F. A . 2004 . The hemlock woolly adelgid: a threat to hemlock in Tennessee . The University of Tennessee, Agric. Ext. Serv ., Pub. SP503-G, Knoxville, TN . Havill , N. P. , L. C. Vieira , and S. M. Salom . 2014 . Biology and control of hemlock woolly adelgid . United States Department of Agriculture, Forest Service , Washington, DC . Howel , W. M. , and R. L. Jenkins . 2004 . Spiders of the eastern United States: a photographic guide . Pearson Education , New York, NY . Jetton , R. M. , F. L. Hasting , and F. P. Hain . 2009 . Arthropod diversity associated with old and secondary growth eastern hemlock (Tsuga canadensis) in the Great Smoky Mountains National Park . J. North Carolina Acad. Sci . 125 : 70 – 77 . Joseph , S. V. , J. L. Hanula , and S. K. Braman . 2011 . Distribution and abundance of Adelges tsugae (Hemiptera: Adelgidae) within hemlock trees . J. Econ. Entomol . 104 : 1918 – 1927 . Google Scholar CrossRef Search ADS PubMed Kizlinski , M. L. , D. A. Orwig , R. C. Cobb , and D. R. Foster . 2002 . Direct and indirect ecosystem consequences of an invasive pest on forests dominated by eastern hemlock . J. Biogeogr . 29 : 1489 – 1503 . Google Scholar CrossRef Search ADS Lynch , C. , P. Lambdin , J. Grant , R. Reardon , and R. Rhea . 2006 . Survey for established predators of elongate hemlock scale in Tennessee . Fla. Entomol . 89 : 527 – 528 . Google Scholar CrossRef Search ADS Mallis , R. E. , and L. K. Rieske . 2010 . Web orientation and prey resources for web-building spiders in eastern hemlock . Environ. Entomol . 39 : 1466 – 1472 . Google Scholar CrossRef Search ADS PubMed Mallis , R. E. , and L. K. Rieske . 2011 . Arboreal spiders in eastern hemlock . Environ. Entomol . 40 : 1378 – 1387 . Google Scholar CrossRef Search ADS PubMed Markó , V. , B. Keresztes , M. T. Fountain , and J. V. Cross . 2009 . Prey availability, pesticides and the abundance of orchard spider communities . Biol. Control 48 : 115 – 124 . Google Scholar CrossRef Search ADS Mayfield , A. , B. Reynolds , C. Coots , N. Havill , C. Brownie , A. Tait , J. Hanula , S. Joseph , and A. Galloway . 2015 . Establishment, hybridization and impact of Laricobius predators on insecticide-treated hemlocks: exploring integrated management of the hemlock woolly adelgid . Forest Ecol. Manag . 335 : 1 – 10 . Google Scholar CrossRef Search ADS McClure , M. S . 1987 . Biology and control of hemlock woolly adelgid . Connecticut Agricultural Experiment Station Bulletin 851 , New Haven, CT . McClure , M. S . 1991 . Density-dependent feedback and population cycles in Adelges tsugae (Homoptera: Adelgidae) on Tsuga canadensis . Environ. Entomol . 20 : 258 – 264 . Google Scholar CrossRef Search ADS McClure , M. S. , S. M. Salom , and K. S. Shields . 2001 . Hemlock woolly adelgid . United States Department of Agriculture, Forest Service, For. Health Technol. Ent. Team, FHTET-2001–03 , Morgantown, WV . Meissle , M. , and A. Lang . 2005 . Comparing methods to evaluate the effects of Bt maize and insecticide on spider assemblages . Agric. Ecosyst. Environ . 107 : 359 – 370 . Google Scholar CrossRef Search ADS Oguri , H. , T. Yoshida , A. Nakamura , M. Soga , and N. Hijii . 2014 . Vertical stratification of spider assemblages in two conifer plantations in central Japan . J. Arachnol . 42 : 34 – 43 . Google Scholar CrossRef Search ADS Orwig , D. A. , D. R. Foster , and D. L. Mausel . 2002 . Landscape patterns of hemlock decline in New England due to the introduced hemlock woolly adelgid . J. Biogeogr . 29 : 1475 – 1487 . Google Scholar CrossRef Search ADS Orwig , D. A. , R. C. Cobb , A. W. D’Amato , M. L. Kizlinski , and D. R. Foster . 2008 . Multi-year ecosystem response to hemlock woolly adelgid infestation in southern New England forests . Can. J. Forest Res . 38 : 834 – 843 . Google Scholar CrossRef Search ADS Pekar , S . 1999 . Effect of IPM practices and conventional spraying on spider population dynamics in an apple orchard . Agric. Ecosyst. Environ . 73 : 155 – 166 . Google Scholar CrossRef Search ADS Quimby , J. W . 1996 . Value and importance of hemlock ecosystems in the eastern United States , pp. 1 – 8 . In S. M. Salom , T. C. Tigner , and R. C. Reardon (eds.), Proceedings of the First Hemlock Woolly Adelgid Review , 12 October 1995, Charlottesville, VA . United States Department of Agriculture, Forest Service, For. Health Technol. Ent. Team, FHTET-96-10 , Morgantown, WV . Raupp , M. J. , R. E. Webb , A. Szczepaniec , D. Booth , and R. Ahern . 2004 . Incidence, abundance, and severity of mites on hemlocks following applications of imidacloprid . J. Arboric . 30 : 108 – 113 . Salem , S. A. , and M. M. Matter . 1991 . Relative effects of neem seed oil and deenate on the cotton leaf worm, Spodoptera littoralis Boisd, and the most prevalent predators in cotton fields at Menoufyia governorate . Bull. Fact Sci. Cairo Univ . 42 : 941 – 952 . SAS Institute . 2006 . SAS user’s guide, statistics, version 9.1 . SAS Institute , Cary, NC , 215 pp. Schowalter , T. D . 1995 . Canopy arthropod communities in relation to forest age and alternative harvest practices in western Oregon . Forest Ecol. Manag . 78 : 115 – 125 . Google Scholar CrossRef Search ADS Schowalter , T. D. , and L. M. Ganio . 1998 . Vertical and seasonal variation in canopy arthropod communities in an old-growth conifer forest in southwestern Washington, USA . B. Entomol. Res . 88 : 633 – 640 . Google Scholar CrossRef Search ADS Shaw , E. M. , M. Waddicor , and A. M. Langan . 2006 . Impact of cypermethrin on feeding behaviour and mortality of the spider Pardosa amentata in arenas with artificial ‘vegetation’ . Pest Manag. Sci . 62 : 64 – 68 . Google Scholar CrossRef Search ADS PubMed Souto , D. R. , and K. S. Shields . 2000 . Overview of hemlock health , pp. 76 – 80 . In K. McManus , K. Shields , and D. Souto (eds.), Proceedings: Symposium on sustainable management of hemlock ecosystems in eastern North America , 22–24 June 1999 , Durham, NH . United States Department of Agriculture, Forest Service, Northeastern Research Station Gen. Tech. Rep. NE-267 , Newton Square, PA . Tattar , T. A. , J. A. Dotson , M. S. Ruizzo , and V. B. Steward . 1998 . Translocation of imidacloprid in three tree species when trunk and soil injected . J. Arboric . 24 : 54 – 56 . Ubick , D. , P. Paquin , P. E. Cushing , and V. Roth (eds). 2005 . Spiders of North America: an identification manual . Am. Arach. Soc . 377 , Poughkeepsie, NY. Vickerman , G. P. , and K. D. Sunderland . 1977 . Some effects of dimethoate on arthropods in winter wheat . J. Appl. Ecol . 14 : 767 – 777 . Google Scholar CrossRef Search ADS Wallace , M. S. , and F. P. Hain . 2000 . Field survey and evaluation of native and established predators of the hemlock woolly adelgid (Homoptera: Adelgidae) in the southeastern United States . J. Environ. Entomol . 29 : 638 – 644 . Google Scholar CrossRef Search ADS Ward , J. S. , M. E. Montgomery , C. A. S.-J. Cheah , B. P. Onken , and R. S. Cowles . 2004 . Eastern hemlock forests: guidelines to minimize the impacts of hemlock woolly adelgid . United States Department of Agriculture, Forest Service, Northeastern Area State and Private Forestry, NA-TP-03-04 , Morgantown, WV . Welch , H. , and G. Haddow . 1993 . The world checklist of conifers . Landsman’s Bookshop Ltd ., Herefordshire, UK . Wu , H. , Y. Luo , J. Shi , W. Chen , X. Yan , and P. Jiang . 1989 . Composition and structure of canopy arthropods in Pinus thunbergii . J. Zhejiang For. College . 27 : 501 – 506 . Published by Oxford University Press on behalf of Entomological Society of America 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US.

Journal

Environmental EntomologyOxford University Press

Published: Aug 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off