Northeastern Naturalist Vol. 23, No. 4 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 555 2016 NORTHEASTERN NATURALIST 23(4):555–570 Influences of a Tsuga canadensis (L.) Carriere (Eastern Hemlock) Riparian Habitat on a Lotic Benthic Community Paige M. Kleindl1, Fred D. Tucker1, Michael G. Commons2, Robert G. Verb1, and Leslie A. Riley1,* Abstract - Tsuga canadensis (Eastern Hemlock) forests provide unique riparian zones that can influence adjacent streams, but increasing mortality from the invasive Adelges tsugae (Hemlock Woolly Adelgid) is eliminating this distinctive landscape component in some regions. The objective of this study was to determine if a stream section within a hemlock ravine harbored a unique benthic community as compared to other sections of the stream that could be threatened in the event of a Hemlock Woolly Adelgid outbreak. We sampled benthic algae and macroinvertebrates in an unnamed tributary of Sugar Creek within Beach City Wildlife Area, OH, in April and September 2015. The stream flows through 3 riparian habitats: beech–maple upland forest, hemlock ravine, and lowland forest dominated by Acer saccharinum (Silver Maple), A. negundo (Box Elder), and Platanus occidentalis (American Sycamore). Our results show that Chironomidae, Navicula, Caloneis, and Nitzschia were the dominant taxa across all 3 stream sections, but that benthic macroinvertebrate richness and density were significantly lower in the hemlock ravine when compared to the lowland habitat. Periphyton community metrics were not significantly affected by riparian habitat. Overall, seasonality was more influential than riparian habitat on benthic community composition; specific taxa were indicative of either the spring or summer season. Connectivity between stream sites and/or the abundance of sandstone bedrock substrate at many sample locations might account for the similarity in benthic communities across these 3 habitats. Introduction Terrestrial and aquatic ecosystems are intimately linked through physical processes and fluxes of energy and nutrients across the riparian ecotone (Gregory et al. 1991, Verry et al. 2000). Riparian areas serve as buffer zones that play key roles in maintaining water temperature, nutrient concentration, and food availability within aquatic systems (Baxter et al. 2005, Gregory et al. 1991, Karr and Schlosser 1978, Naiman and Décamps 1997, Richardson and Danehy 2007). Riparian zones can be particularly influential on forested headwater streams, where light availability is often a limiting resource and allochthonous energy subsidies can be substantial (Bilby and Bisson 1992, Hill and Knight 1988). Dense, low, overhanging canopies greatly reduce light intensity at the water’s surface, but high, relatively open canopies allow more light to reach the stream (Giller and Malmqvist 1998, Gregory et al. 1991). Increases in light availability stimulate in-stream primary production and provide autochthonous resources for macroinvertebrates (Willacker et al. 2009). In 1Department of Biological and Allied Health Sciences, Ohio Northern University, Ada, OH 45810. 2Department of History, Politics, and Justice, Ohio Northern University, Ada, OH 45810. *Corresponding author - l-riley.1@onu.edu. Manuscript Editor: David Orwig Northeastern Naturalist 556 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 Vol. 23, No. 4 addition, allochthonous energy subsidies enter the stream in the form of leaf litter and terrestrial invertebrates (Bilby and Bisson 1992). Small streams can be largely dependent on energy subsidies from the surrounding forest, which can influence the distribution of macroinvertebrates in a stream (Flory and Milner 1999, Vinos 2001). The conifer Tsuga canadensis (L.) Carriere (Eastern Hemlock) creates a unique riparian zone. Eastern Hemlock has a high leaf-area index that increases shading year-round and creates a cool, moist, forest understory that provides thermal stability to the air, soil, and water beneath its dense canopy (Dayton 1972, Godman and Hadley 2000, Lancaster 1990, Snyder et al. 2002). Eastern Hemlock also regulates nutrient cycling, contributes a consistent amount of leaf litter to adjacent streams throughout the growing season, and stabilizes stream base-flows due to persistent and elevated transpiration rates in spring and fall (Adkins and Rieske 2014, Ellison et al. 2005, Ford and Vose 2007, Nuckolls et al. 2009, Webster et al. 2012, Welsh and Droege 2001, Yorks et al. 2000). Eastern Hemlock stands may also constrain food resources in streams by shading periphyton communities (Rowell and Sobczak 2008) and providing low-quality leaf litter for stream consumers (Maloney and Lamberti 1995, Strohm 2014). Eastern Hemlock needles decay more slowly and support fewer macroinvertebrates than leaves of many deciduous riparian plant species (Maloney and Lamberti 1995, Strohm 2014). Eastern Hemlock-dominated riparian zones also have lower macroinvertebrate abundance and different community compositions than streams with deciduous riparian areas (Snyder et al. 2002, Willacker et al. 1999). Eastern Hemlock forests have declined substantially in the last 2 decades (Evans et al. 2011, Orwig et al. 2002). Widespread defoliation and mortality of Eastern Hemlocks have largely been attributed to Adelges tsugae Annand (Hemlock Woolly Adelgid, hereafter HWA), a small, piercing and sucking insect native to East Asia that feeds on a number of hemlock species (McClure 1991). HWA is rapidly spreading throughout the eastern US, and once infested, Eastern Hemlock stands can suffer complete mortality within 5 years (McClure 1991). Recent studies suggest that hemlock regeneration following infestation is largely absent, and no infested tree or stand has been found to exhibit any sign of recovery (Orwig and Foster 1998). Our primary objective for this study was to gather baseline stream benthic community data in an HWA-free Eastern Hemlock riparian forest. Previous studies focused on the relationship between hemlock forests and macroinvertebrate communities, not including the important algal component of the benthic community (e.g., Snyder et al. 2002, Willacker et al. 2009). In this study, we investigated whether the benthic community in a hemlock ravine was different from the benthic communities in neighboring sections of the stream with deciduou s riparian zones. Field-site Description Beach City Wildlife Area is a 773-ha state wildlife area located in Tuscarawas County in eastern Ohio (Ohio Department of Natural Resources, Division of Wildlife 2015), within the unglaciated Allegheny Plateau (Fenneman 1938). This area is Northeastern Naturalist Vol. 23, No. 4 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 557 also located within the 945.86-km2 Sugar Creek watershed (Ohio EPA, Division of Surface Water 2005). The region contains Pennsylvanian-era bedrock with extensive sections of eroded Massillon sandstone remnants and slump blocks exposed by glacial outwash (Camp 2006). Elevation ranges from 274 m to 341 m due to sandstone weathering and the creation of ravines and cliffs at some locations. The soil consists primarily of moderately drained silt loams (USDA, NRCS 2015). Our study sites were located along an unnamed 1st-order tributary stream of Sugar Creek that traverses 3 distinct microclimates in the Beach City Wildlife Area (Fig. 1): (1) upland deciduous forest dominated by Fagus grandifolia Ehrh (American Beech) and Acer saccharum Marshall (Sugar Maple), (2) glacial refugia in a sandstone ravine dominated by Eastern Hemlock and Betula alleghaniensis Britt. (Yellow Birch), and (3) lowland riparian forest dominated by Salix spp. (willow), Acer saccharinum L. (Silver Maple), and Platanus occidentalis L. (American Sycamore). Habitat zones were 25–75 m wide on either side of the stream throughout the study area. Materials and Methods Field sampling We sampled 13 sites in the Beach City stream on 26 April 2015 and 1 September 2015. All sites were separated by a distance of 100 m, with the exception of sites 8 and 9, which were separated by 40 m (Fig. 1). Site 9 was located at the top of a waterfall, and marked the beginning of the hemlock ravine. Site 8 was the plunge pool at the base of the waterfall. At each site, physical measurements included wetted-stream width, current velocity, and water depth. We calculated current Figure 1. Beach City Wildlife Area Tuscarawas County, OH. (A) Generalized location of Beach City Wildlife Area (signified by the star) located within the Unglaciated Allegheny Plateau of Ohio. (B) Aerial photograph of Beach City Wildlife Area with the 13 designated sampling sites identified with black dots and the 3 habitat type s labeled accordingly. Northeastern Naturalist 558 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 Vol. 23, No. 4 velocity as the time it took a fishing bobber to travel 1 m (average of 3 trials), and measured water depth with meter sticks at 5 locations within the riffle habitat. We used a YSI 556 portable water-quality meter (Xylem Inc., Yellow Springs, OH) to measure temperature, specific conductance, total dissolved solids, pH, salinity, and dissolved oxygen. Stream water was collected in 500-ml bottles and placed on ice for laboratory analyses of alkalinity and turbidity. At each site, we selected riffle habitat closest to the point marking 100 m from the previous sample. For periphyton (epilithic algae) collections, we used a random number generator to choose 5 rocks from each transect and scraped a 5.0-cm2 area on each rock using a rigid, sterile O-ring and stiff toothbrush. At sites with bedrock substrates, we employed a Loeb periphyton sampler (Loeb 1981) to extract the periphyton samples. We rinsed recovered periphyton material with stream water, collected a composite sample in a 50-mL Falcon® tube, and preserved each in 15% buffered formalin. We used a Surber sampler (area = 225 cm2, mesh size = 500 μm) to collect macroinvertebrates from 3 locations within the riffle habitat, which we combined into 1 composite sample and preserved in 70% ethanol. We assessed percent cover of macroalgae and aquatic mosses at each riffle (Sheath et al. 1986), collected voucher samples, and preserved them in 70% ethanol. To assess the makeup of the stream substrate in riffle habitat, we set a 0.25 m x 0.25 m grid to the right of the second Surber sampler and measured and recorded percent cover and sediment-particle size (Udden-Wentworth scale). We noted the presence of continuous sheets of bedrock that dominated some sample sites. We employed the point-quarter technique to survey woody plants along the stream and identified them to species (Cottam et al. 1953). The understory light environment was characterized using hemispherical photography (Robison and McCarthy 1999). We took photographs of the canopy 1 m above the center of each riffle using an Olympus 8-mm lens on an Olympus E-510 digital SLR camera. We used GLA software (ver. 2.0; Frazer et al. 1999) to analyze the digital images and determine percent canopy cover. Laboratory analyses We used a HANNA HI 4811 Test Kit (Hanna Instruments, Woonsocket, RI) and a Hach 2100P turbidity meter (Hach, Loveland, CO) to determine total alkalinity and turbidity, respectively, for each water sample. We placed a small, homogenized quantity of each periphyton sample in a Palmer-Maloney counting chamber to (1) enumerate and identify 300 soft-bodied algal cells, and (2) survey the abundance and condition of diatoms. When possible, we identified non-diatom (soft-bodied) algae to genus using taxonomic references including Dillard (1999), Prescott (1962), and Whitford and Schumacher (1984), and updated taxonomy from Wehr et al. (2015). To identity diatoms to genus, we boiled a subsample (10 ml) from each collection in 35% hydrogen peroxide for 60 minutes, conducted a series of distilled-water dilutions to remove oxidation byproducts, evaporated the samples onto coverslips, and mounted them on microscope slides using the mounting medium ZraxTM. We identified 300 diatom valves Northeastern Naturalist Vol. 23, No. 4 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 559 to genus enumerated along a transect(s) using a Meiji MX4300L brightfield light microscope (na = 1.30). When the densities were low, we limited counts to eight 18-mm transects. Identification of diatom genera was based on taxonomic literature including work from the US (Krammer and Lange-Bertalot 1986, 1988, 1991a, 1991b; Patrick and Reimer 1966, 1975; Spaulding et al. 2010). We examined macroalgae and aquatic mosses under a Meiji SZH-ILLDTM stereoscope and used field notes to separate the composite samples into discrete entities for further resolution using a compound microscope. Each specimen was examined using a Meiji BX40TM microscope and identified to genus, using Crum (1983) and Dillard (1990, 1991a, 1991b, 1993, 1999), Prescott (1962), Taft and Taft (1971), and Whitford and Schumacher (1984). We studied macroinvertebrates with a Meiji Techno EMZ-13 stereoscope. Insects were identified to family, other arthropods to order (e.g., Amphipoda, Collembola, and Decapoda), Oligochaeta to subclass, and Nematoda to phylum (Merritt et al. 2008, Thorp and Covich 2010, Voshell 2002). Statistical analyses To analyze the fixed effects of habitat (upland vs. ravine vs. lowland) and season (April vs. September), we used a multivariate analysis of variance (MANOVA). Response variables included biotic community estimates (periphyton and macroinvertebrate communities: abundance, Shannon diversity, taxa richness). We employed probability plots and Anderson-Darling tests to assess normality of the response variables and determined that it was necessary to square-root tranform both periphyton and macroinvertebrate density to normalize the data. We conducted analyses of variance (ANOVAs) to determine the significance of individual response-variables, and Bonferroni-Dunn multiple comparison tests to examine significant differences between habitat types. Analyses were performed in Minitab 17.2.1 (Minitab Inc., 2013). We initially employed detrended correspondence analysis (DCA) to determine if the variation in riffle community structure warranted the use of canonical correspondence analysis (CCA, unimodal response) or redundancy analysis (RDA, linear response). The DCA results indicated that the gradient length of the first axis was <3 standard deviations; thus, RDA was used to analyze benthic community structure and environmental parameters. To reduce the impact of autocorrelation, we checked the 14 environmental parameters for high correlation coefficients (r > 0.85) and variance-inflation factors with values ≥ 10 (Pan et al. 1996, ter Braak and Šmilauer 1998). The significance of the first RDA axis was tested using Monte Carlo permutation tests (1000 random permutations, α = 0.05). We analyzed all datasets separately using a series of nonparametric multivariate analyses including multi-response permutation procedure (MRPP), indicator species analysis (ISA), and nonmetric multidimensional scaling (NMDS). The data matrices were standardized prior to analyses, which were conducted in the software PC-ORD (version 6.19, MjM Software Design). We employed MRPP to test for differences between sites arranged according to habitat (upland vs. hemlock ravine vs. lowland). Northeastern Naturalist 560 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 Vol. 23, No. 4 Default software settings were used, including a Euclidean distance measure. When results of MRPP indicated significant differences, we ran a follow-up ISA to describe differences in community structure detected in the original MRPP. ISA was used to determine the relative degree of exclusivity or affinity that certain taxa or groups may have for a particular regional group. We interpreted ISA results using indicator values tested against a randomized chance-indicator value. To visualize these differences among sites in ordination space, we conducted a follow up NMDS to ordinate each of the matrices found to be significant in the MRPP/ISA analyses. Results Taxa survey We collected 41 diatom genera and 18 genera of soft-bodied algae. The mostdominant diatom taxa included Navicula, Nitzschia, Gomphonema, and Caloneis. The most abundant soft-bodied algae were Pseudanabaena and Oscillatoria. We also collected 47 macroinvertebrate taxa. The most-dominant insect families were the Chironomidae, Perlodidae, Hydropsychidae, and Tipulidae. We recorded 10 genera of macroalgae and aquatic mosses; Cladophora and Fontinalis were the most abundant macroalgae and aquatic moss taxa, respectively. We documented 18 species of woody plants, including the species closest to the sampling sites: Sugar Maple, Acer rubrum L. (Red Maple), Yellow Birch, and American Sycamore. Benthic abundance, richness and diversity Overall, seasonality significantly influenced the benthic community (MANOVA: F = 3.39, P = 0.012, Wilk’s λ = 0.419), but habitat type did not (F = 1.94, P = 0.065, Wilk’s λ = 0.353). With respect to individual response variables, periphyton richness (season: P = 0.297, habitat: P = 0.986, season x habitat: P = 0.664; Fig. 2A), periphyton diversity (season: P = 0.696, habitat: P = 0.506, season x habitat: P = 0.638; Fig. 2B), and, periphyton density (season: P = 0.495, habitat: P = 0.426, season x habitat: P = 0.908; Fig. 2C) were not significantly affected by habitat or season. Conversely, macroinvertebrate density (P = 0.027), macroinvertebrate richness (P = 0.018), and macroinvertebrate diversity (P = 0.016) were all significantly affected by habitat type (Fig. 3). Only macroinvertebrate diversity differed across seasons, with greater diversity in September compared to April (P = 0.004; Fig. 3B). Interactions between habitat and season were not significant (density: P = 0.994, richness: P = 0.870, diversity: P = 0.593). Macroinvertebrate density, richness, and diversity were lower in the hemlock ravine, and post-hoc comparisons revealed that these values were significantly different from those for the lowland habitat with respect to richness and density, regardless of season (Fig. 3). Multivariate analyses of community structure We removed soft-bodied algae from the multivariate analyses because of the widespread distribution of filamentous cyanobacteria. Consistently high cell-counts of Pseudanabaena skewed the results to depict a homogenous benthic community at all sites. The RDA analysis, based on diatoms and macroinvertebrates, was strongly Northeastern Naturalist Vol. 23, No. 4 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 561 influenced by the environmental variables correlated with the fir st ordination axis: alkalinity, specific conductance, temperature, and canopy cover (Table 1). These predominant variables all showed strong seasonal fluctuation (Table 2). There is a visible spatial split between most of the habitat types based on seasonality along the first axis (Fig. 4). Spring habitat types were skewed to the right due to lower water temperatures and a more open canopy, while the summer habitat types were skewed to the left due to higher water temperatures and denser canopy coverage (Fig. 4). Figure 2. (A) Periphyton genera richness, (B) periphyton diversity (H'), and (C) periphyton cell density (mm-2). Boxes represent the 25th and 75th percentiles, and whiskers represent the 5th and 95th percentiles. Lines in boxes represent median values. Northeastern Naturalist 562 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 Vol. 23, No. 4 The Monte Carlo permutation test showed significance (P = 0.05) across the 3 axes that explained 40.2% of the taxa variance (Table 1). We examined the diatom and macroinvertebrate communities within the habitat zones using NMDS and MRPP. The MRPP based on habitat was significant (P = 0.021); however, the A-value was marginal (A = 0.054), indicating similarity between the dominant taxa at each location (Fig. 5). In both spring and summer, Navicula and Chironomidae were the dominant taxa in all 3 habitat types, but there were also some indicator taxa for each habitat. Hydropsychidae and Hydracarina were charactersitic of the upland habitat, Chironomidae was an indicator Figure 3. (A) Macroinvertebrate richness, (B) macroinvertebrate diversity (H'), and (C) macroinvertebrate density (m-2). Boxes represent the 25th and 75th percentiles, and whiskers represent the 5th and 95th percentiles. Lines in boxes represent median values. Boxes with the same letters are not significantly different (P less than 0.05) as determined by Bonferroni- Dunn post-hoc comparison tests. Northeastern Naturalist Vol. 23, No. 4 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 563 for the ravine habitat, and Navicula, Perlodidae, and Elmidae were all indicative of the lowland habitat. The MRPP based on seasonality was significant (P < 0.001) and was influenced by a number of taxa that showed strong patterns of seasonality (Fig. 5). The spring-season indicator taxa were Chironomidae, Odontoceridae, Achnanthidium, Nemouridae, Craticula, Fragilaria, Caloneis, and Surirella. The summer indicator taxa were Cocconeis, Amphora, Hydropsychidae, Nitzschia, Tryblionella, Veliidae, and Gyrosigma. Table 1. RDA summary table: λ = eigenvalue, S = percent variance explained by the corresponding axis, TVE = total variance explained, and r = correlation coefficient between axis and influential environmental parameters. All 3 axes were statistically significant (P < 0.05) as determined by the Monte Carlo permutation test. Axis λ S Environmental parameter (r) I 13.6114 17.5 Alkalinity (-0.829), specific conductance (-0.794), temperture (-0.768), canopy cover (0.801) II 9.123 11.7 Width (0.331), nearest tree (-0.385) III 8.598 11.0 pH (0.769) TVE 40.2 Figure 4. Diatom- and macroinvertebrate-based redundancy analysis (RDA) at Beach City Wildlife Area with environmental variables represented by arrows. Cond = specific conductance, Can = canopy cover, MA = macroalgal and moss coverage, D = stream depth, W = stream width, T = turbidity, Alk = alkalinity, and 1st tree = distance to nearest tree. The dashed line indicates the split between the spring and summer months. Monte Carlo permutation tests revealed the first axis of the biplot to be signific ant (P < 0.05). Northeastern Naturalist 564 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 Vol. 23, No. 4 Table 2. Summary of descriptive statistics for selected physical, chemical, dominant riparian tree species and habitat variables (mean value with ranges in parentheses) for sites sampled at Beach City Wildlife Area. Beach City Habitat Types Riparian forest (sites 1–5) Hemlock forest (glacial refugia, sites 6–8) Beech–maple forest (sites 9–13) Variable Spring Fall Spring Fall Spring Fall Sp. conductance (μS/cm) 1159.20 1417.60 1276.67 1656.00 1304.60 1703.00 (1116.00–1205.00) (1314.00–1537.00) (1267.00–1283.00) (1625.00–1678.00) (1286.00–1326.00) (1691.00–1714.00) Current velocity (cm/s) 3.25 4.22 4.52 4.05 3.69 3.70 (2.38–3.77) (2.77–7.58) (2.63–5.49) (3.45–4.75) (1.43–8.27) (1.08–5.54) Dissolved oxygen (mg/L) 12.89 10.96 10.79 13.09 11.45 11.51 (11.06–14.53) (10.49–11.89) (10.50-11.12) (11.84–15.22) (11.02–11.71) (10.94–12.56) Channel width (m) 3.27 2.20 4.42 2.61 2.88 2.06 (1.70–5.80) (1.35–2.80) (2.90–6.85) (1.63–3.90) (1.20–6.30) (0.86–3.48) Open canopy (%) 79.08 18.22 45.93 13.16 73.57 19.60 (69.43–79.29) (12.00–27.51) (38.95–55.88) (9.38–19.86) (54.94–82.90) (13.12–24.46) pH 8.09 7.82 8.38 8.22 8.29 8.38 (7.82–8.21) (7.27–8.08) (8.37–8.40) (7.97–8.36) (8.20–8.40) (8.35–8.43) Temperature (°C) 12.86 18.43 13.97 19.83 13.92 21.72 (12.13–13.17) (15.71–20.47) (12.91–14.54) (19.03–20.50) (13.62–14.58) (21.21–22.14) Thalweg depth (cm) 7.07 5.01 8.80 3.72 6.92 5.06 (5.40–9.00) (3.46–7.28) (7.00–10.40) (3.14–4.46) (3.80–10.40) (2.63–6.70) Total alkalinity (mg/L) 140.00 196.00 133.33 220.00 148.00 228.00 (120.00–160.00) (180.00–200.00) (120.00–140.00) (220.00–220.00) (140.00–160.00) (220.00–240.00) Turbidity (NTU) 9.90 1.91 11.50 2.90 9.38 3.85 (6.21–13.10) (1.08–2.99) (8.31–13.50) (1.21–6.23) (3.24–27.20) (1.43–9.15) Stream gradient (m/km) 9.90 40.63 21.31 Dominant riparian tree(s) m2/hectare Ulmus rubra 73.47 Tsuga canadensis 38.85, Betula alleghaniensis 34.54 Acer saccharum 78.23 Riffle substrate (%) Cobble 52.12 (39.00–88.00) Sandstone Bedrock 100.00 (100.00–100.00) Cobble 72.26 (58.00–95.00) Northeastern Naturalist Vol. 23, No. 4 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 565 Discussion Seasonal variation influenced the stream benthic community much more than the spatial change in riparian habitat; all 3 habitats displayed seasonal differences in benthic community composition. Various temporal factors, such as canopy cover and temperature, could account for much of the seasonal variation. Although the hemlock ravine did not have a unique benthic community when compared to adjacent stream sections, the hemlock benthic community did have significantly lower macroinvertebrate richness and density than the lowland habitat . The periphyton community was generally unaffected by the change in riparian habitat along the stream. We found that hardy generalist taxa—Navicula, Nitzschia, and Caloneis—were dominant throughout the stream. These findings were expected because all 3 of these genera are known to be widespread and tolerant to environmental changes within North American freshwater communities (Wehr and Sheath 2003). Conversely, habitat did affect the benthic macroinvertebrate community, especially within the hemlock ravine. Macroinvertebrate richness and density were significantly lower in the ravine than in the other 2 habitat zones. The substantial amount of bedrock substrate in this habitat zone is a possible explanation for the lower richness and density we observed. Bedrock is an unsuitable substrate for many macroinvertebrates, with abundance decreasing on substrate sizes larger than cobbles (Jowett and Richardson 1990). Another explanation for community similarity across riparian zones could be the level of taxonomic identification we employed. Groups with high species richness, such as Chironomidae and Navicula, may contain species that were overlooked in this study . Figure 5. Diatom and macroinvertebrate NMDS scatterplot at Beach City Wildlife Area with site distributions based on season and habitat type. The largest differences were between season (MRPP A = 0.09; P < 0.001), while stream habitat type (MRPP A = 0.054; P = 0.021) showed smaller differences. Northeastern Naturalist 566 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 Vol. 23, No. 4 Although we expected the hemlock ravine to have a distinctly different benthic community, much like its unique terrestrial flora (Webster et al. 2012), we found that it did not. In contrast, Willacker et al. (2009) found a distinct benthic community within an isolated hemlock forest stream. However, we sampled a stream continuum that passed through the beech–maple forest prior to the hemlock ravine. This continuum may not have included sufficient area along the stream to provide a meaningful difference in algal and macroinvertebrate community structure. In addition, connectivity and dispersal of taxa throughout the stream segments may have played a role in community similarity. The limited taxonomic responses to terrestrial habitat were restricted to the riffle habitats in the lowland region. Cocconeis was, unsurprisingly, an indicator for lowland habitat; this diatom is often an epiphytic taxon that attaches to aquatic mosses and macroalgae (Blindow 1987), both of which were more abundant in the lowland riffle habitats. Riffle beetles (Elmidae) were also an indicator for this stream segment that contained a high percentage of cobble substrates within riffle habitats (Elliot 2008). We identified distinct indicator taxa for both seasons. Water temperature and canopy-cover fluctuations from spring to summer months most likely influenced both periphyton and macroinvertebrate indicator taxa (Banks et al. 2007, Hill et al. 2009). For example, macroinvertebrate larval development, emergence rates from the stream, and mating schedules are all influenced by season, and can also vary greatly among families (Banks et al. 2007). We found that the hemlock ravine had lower macroinvertebrate density, corroborating findings from previous studies (e.g., Snyder et al. 2002, Willacker et al. 2009). However, in contrast, we found that macroinvertebrate richness was lower in the hemlock ravine (see Ellison et al. 2005, Snyder et al. 2002 for contrasting findings). We also found that the hemlock ravine did not have a distinct benthic community when compared to adjacent stream segments. This finding contradicts those of previous studies (e.g., Ellison et al. 2005, Snyder et al. 2002, Willacker et al. 2009) and could be due to the limited spatial extent of our study and high degree of connectivity between stream segments. Finally, while season significantly affected periphyton and macroinvertebrate communities, the response was similar across stream segments. This contradicts previous findings for macroinvertebrates where some shredder taxa were more abundant in hemlock streams during summer compared to deciduous streams (Adkins and Riese 2014). HWA has not yet invaded the Eastern Hemlock forest within Beach City Wildlife Area; thus, the results in this study provide an important spatial and temporal baseline dataset to describe the benthic community in this stream in light of a likely future invasion. Acknowledgments We thank The Ohio Department of Natural Resources Division of Wildlife for access to the field site. We are grateful for field assistance from participants in the 2015 Ohio Northern University Field Semester: Janet Deardorff, Stephanie Estell, Nicole Berry, Emily Hennemen, and Jonathan Stechschulte. Lastly, we thank Melissa Hall and Jay Mager for editorial assistance, and Sharyn and Mike Zembower at the Ohio Northern University Metzger Nature Center for accommodations and help with field equ ipment. Northeastern Naturalist Vol. 23, No. 4 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 567 Literature Cited Adkins, J.K., and L.K. Riese. 2014. A terrestrial invader threatens a benthic community: Potential effects of Hemlock Woolly Adelgid-induced loss of Eastern Hemlock on invertebrate shredders in headwater streams. Biological Invasions 17 :116 –1179. Banks, J.L., J. Li, and A.T. Herlihy. 2007. Influence of clearcut logging, flow duration, and season on emergent aquatic insects in headwater streams of the Central Oregon Coast Range. Journal of the North American Benthological Society 26(4):620–632. Baxter, C.V., K.D. Fausch, and W. Carl Saunders. 2005. Tangled webs: Reciprocal flows of invertebrate-prey link streams and riparian zones. Freshwater B iology 50(2):201–220. Bilby, R.E., and P.A. Bisson. 1992. Allochthonous versus autochthonous organic matter contributions to the trophic support of fish populations in clear-cut and old-growth forested streams. Canadian Journal of Fisheries and Aquatic Sciences 49(3):540–551. Blindow, I. 1987. The composition and density of epiphyton on several species of submerged macrophytes: The neutral-substrate hypothesis tested. Aquatic Botany 29(2):157–168. Camp, M.J. 2006. Roadside Geology of Ohio. Mountain Press Publishing, Missoula, MT. 416 pp. Cottam, G., J.T. Curtis, and B.W. Hale. 1953. Some sampling characteristics of a population of randomly dispersed individuals. Ecology 34(4):741–757. Crum, H.A. 1983. Mosses of the Great Lakes Forest (3rd Edition). University Herbarium, University of Michigan, Ann Arbor, MI. 417 pp. Dayton, P.K. 1972. Toward an understanding of community resilience and the potential effects of enrichments to the benthos at McMurdo Sound, Antarctica. Pp. 81–96, In B.C. Parker (Ed.). Proceedings of the Colloquium on Conservation Problems in Antarctica. Allen Press, Lawrence, KS. 356 pp. Dillard, G.E. 1990. Freshwater Algae of the Southeastern United States: Chlorophyceae: Zygnematales: Zygnemataceae, Mesotaeniaceae and Desmidiaceae, J. Cramer, Berlin, Germany. 276 pp. Dillard, G.E. 1991a. Freshwater Algae of the Southeastern United States, Part 5 Section 3: Chlorophyceae, Zygnematales, Desmidiaceae, J. Cramer, Berlin, Germany. 310 pp. Dillard, G.E. 1991b. Freshwater Algae of the Southeastern United States: Chlorophyceae: Zygnematales: Desmidiaceae, J. Cramer, Berlin, Germany. 231 pp. Dillard, G.E. 1993, Freshwater Algae of the Southeastern United States: Chlorophyceace: Zygnematales: Desmidiaceae, J. Cramer, Berlin, Germany. 166 pp. Dillard, G.E. 1999. Common Freshwater Algae of the United States: An Illustrated Key to the Genera (Excluding the Diatoms), Gebrüder Borntraeger, Berlin, Germany. 173 pp. Dodds, W.K. 2002. Freshwater Ecology: Concepts and Environmental Applications. Academic Press, Waltham, MA. 569 pp. Elliott, J.M. 2008. The ecology of riffle beetles (Coleoptera: Elmidae). Freshwater Reviews 1:189–203. Ellison, A.M., M S. Bank, B.D. Clinton, E.A. Colburn, K. Elliott, C.R. Ford, D.R. Foster, B.D. Kloeppel, J.D. Knoepp, G.M. Lovett, J. Mohan, D.A. Orwig, N.L. Rodenhouse, W.V. Sobczak, K.A. Stinson, J.K. Stone, C.M. Swan, J. Thompson, B. Von Holle, and J.R. Webster. 2005. Loss of foundation species: Consequences for the structure and dynamics of forested ecosystems. Frontiers in Ecology and the Environment 3(9):479–486. Evans, D.M., W.M. Aust, C.A. Dolloff, B.S. Templeton, and J.A. Peterson. 2011. Eastern Hemlock decline in riparian areas from Maine to Alabama. Northern Journal of Applied Forestry 28(2):97–104. Northeastern Naturalist 568 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 Vol. 23, No. 4 Fenneman, N.M. 1938. Physiography of Eastern United States. McGraw-Hill Book Company, New York, NY. 714 pp. Flory, E.A., and A.M. Milner. 1999. Influence of riparian vegetation on invertebrate assemblages in a recently formed stream in Glacier Bay National Park, Alaska. Journal of the North American Benthological Society: 18:261–273. Ford, C.R., and J.M. Vose. 2007. Tsuga canadensis (L.) Carr. mortality will impact hydrologic processes in southern Appalachian forest ecosystems. Ecological Applications 17(4):1156–1167. Frazer, G.W., C.D. Canham, and K.P. Lertzman. 1999. Gap Light Analyzer. Vers. 2.0. Cary Institute of Ecosystem Studies, Millbrook, NY. Giller, P.S., and B. Malmqvist. 1998. The Biology of Streams and Rivers. Oxford University Press, Oxford, UK. 272 pp. Godman, R.M., and K. Lancaster. 1990. Tsuga canadensis (L.) Carr. Eastern Hemlock. Silvics of North America 1(1):604–612. Gregory, S.V., F.J. Swanson, W.A. McKee, and K.W. Cummins. 1991. An ecosystem perspective of riparian zones. BioScience 41(8)540–551. Hadley, J.L. 2000. Understory microclimate and photosynthetic response of saplings in an old-growth Eastern Hemlock (Tsuga canadensis L.) forest. Ecoscience 7(1):66–72. Hill, W.R., and A.W. Knight. 1988. Nutrient and light limitation of algae in two northern California streams. Journal of Phycology 24(2):125–132. Jowett, I.G., and J. Richardson. 1990. Microhabitat preferences of benthic invertebrates in a New Zealand river and the development of in-stream flow habitat-models for Deleatidium spp. New Zealand Journal of Marine and Freshwater Research 24(1 ):19–30. Karr, J.R., and J. Schlosser. 1978. Water resources and the land–water interface. Science 201(4352):229–234. Krammer, K., and H. Lange-Bertalot. 1986. Bacillariophyceae. 1. Teil: Naviculaceae, VEB Gustav Fisher Verlag, Jena, Germany. 876 pp. Krammer, K., and H. Lange-Bertalot. 1988. Bacillariophyceae. 2. Teil: Epithemiaceae, Bacillariaceae, Surirellaceae, VEB Gustav Fisher Verlag, Jena, Germany. 437 pp. Krammer, K., and H. Lange-Bertalot, H. 1991a. Bacillariophyceae. 3. Teil: Centrales, Fragilariaceae, Eunotiaceae, Achnanthaceae, VEB Gustav Fisher Verlag, Jena, Germany. 596 pp. Krammer, K., and H. Lange-Bertalot. 1991b. Bacillariophyceae. 1. Teil: Achnanthaceae, Kritische Erganzungen zu Navicula (Lineolatae) und Gomphonema, VEB Gustav Fisher Verlag, Jena, Germany. 596 pp. Loeb, S.L. 1981. An in situ method for measuring the primary productivity and standing crop of the epilithic periphyton community in lentic systems. Limnology and Oceanography 26(2):394–399. Maloney, D.C., and G.A. Lamberti. 1995. Rapid decomposition of summer-input leaves in a northern Michigan stream. American Midland Naturalist 133:184–195. McClure, M.S. 1991. Nitrogen fertilization of hemlock increases susceptibility to Hemlock Woolly Adelgid. Journal of Arboriculture 17(8):227–229. Naiman, R.J., and H. Décamps. 1997. The ecology of interfaces: Riparian zones. Annual Review of Ecology and Systematics 28:621–658. Nuckolls, A.E., N. Wurzburger, C.R. Ford, R.L. Hendrick, J.M. Vose, and B.D. Kloeppel. 2009. Hemlock declines rapidly with Hemlock Woolly Adelgid infestation: Impacts on the carbon cycle of southern Appalachian forests. Ecosystems 12(2):179–190. Ohio Department of Natural Resources, Division of Wildlife. 2015. Beach City Wildlife Area. Available online at http://wildlife.ohiodnr.gov/beachcity. Accessed May 2016. Northeastern Naturalist Vol. 23, No. 4 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 569 Ohio Environmental Protection Agency (Ohio EPA), Division of Surface Water. 2005. Recreational-use water quality of the Sugar Creek watershed. Co lumbus, OH. Orwig, D.A., and D.R. Foster. 1998. Forest response to the introduced Hemlock Woolly Adelgid in southern New England, USA. Journal of the Torrey Botanical Society 125(1):60–73. 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. Journal of Biogeography 29:1475–1487. Pan, Y., R.J. Stevenson, B.H. Hill, A.T. Herlihy, and G.B. Collins. 1996. Using diatoms as indicators of ecological conditions in lotic systems: A regional assessment. Journal of the North American Benthological Society 15(4):481–495. Patrick, R., and C.W. Reimer. 1966. The Diatoms of the United States, Volume I. Monograph 13. Academy of Natural Sciences of Philadelphia, Philadelphia, PA. 638 pp. Patrick, R., and C.W. Reimer. 1975. The Diatoms of the United States, Volume II. Monograph 13. Academy of Natural Sciences of Philadelphia, Philadelphia, PA. 688 pp. Prescott, G.W. 1962. Algae of the Western Great Lakes Area. W.M.C. Brown Publishers, Dubuque, IA. 977 pp. Richardson, J.S., and R.J. Danehy. 2007. A synthesis of the ecology of headwater streams and their riparian zones in temperate forests. Forest Science 53(2):131–147. Rios, S.L., and R.C. Bailey. 2006. Relationship between riparian vegetation and stream benthic communities at three spatial scales. Hydrobiologia 553(1):153–160. Robison, S.A., and B.C. McCarthy. 1999. Potential factors affecting the estimation of light availability using hemispherical photography in oak forest understories. Journal of the Torrey Botanical Society 126(4):344–349. Rowell, T.J., and W.V. Sobczak. 2008. Will stream periphyton respond to increases in light following forecasted regional hemlock mortality? Journal of Freshwater Ecology 23(1): 33–40. Sheath, R.G., M.O. Morison, J.E. Korch, D. Kaczmarczyk, and K.M. Cole. 1986. Distribution of stream macroalgae in south-central Alaska. Hydrobiologia 135:259–269. Snyder, C.D., J.A. Young, D.P. Lemarié, and D.R. Smith. 2002. Influence of Eastern Hemlock (Tsuga canadensis) forests on aquatic invertebrate assemblages in headwater streams. Canadian Journal of Fisheries and Aquatic Sciences 59(2):262–275. Spaulding, S.A., D.J. Lubinski, and M. Potapova. 2010. Diatoms of the United States. Available online at http://westerndiatoms.colorado.edu. Accessed May 2016. Strohm, C. 2014. Changing litter resources associated with Hemlock Woolly Adelgid invasion affect benthic communities in headwater streams. M.Sc. Thesis. University of Kentucky. Lexington, KY. Taft, C.E., and C.W. Taft. 1971. The Algae of Western Lake Erie. Ohio Biological Survey Volume 4. 189 pp. ter Braak, C.J.F., and P. Šmilauer. 1998. Canoco reference manual and user’s guide to Canoco for Windows: Software for canonical community ordination, Version 4.0, Microcomputer Power, Ithaca, NY. Thorp, J.H., and A.P. Covich. 2010. Ecology and Classification of North American Freshwater Invertebrates. Academic Press, Waltham, MA. 1021 pp. US Department of Agriculture, Natural Resources Conservation Services (USDA, NRCS). 2015. Custom Soil Report for Tuscarawas County, Ohio. Available online at http:// websoilsurvey.sc.egov.usda.gov/App/HomePage.htm. Verry, E.S., J.W. Hornbeck, and C.A. Dolloff. 1999. Riparian Management in Forests of the Continental Eastern United States. CRC Press, Boca Raton, FL. 3 20 pp. Northeastern Naturalist 570 P.M. Kleindl, F.D. Tucker, M.G. Commons, R.G. Verb, and L.A. Riley 2016 Vol. 23, No. 4 Vinos, C.V. 2001. Riparian leaf-litter processing by benthic macroinvertebrates in a woodland stream of central Chile. Revista Chilena de Historia Natural 74:445–453. Voshell, J.R., Jr. 2002. A Guide to Common Freshwater Invertebrates of North America. The McDonald and Woodward Publishing Company, Granville, OH. 454 pp. Webster, J.R., K. Morkeski, C.A. Wojculewski, B.R. Niederlehner, E.F. Benfield, and K.J. Elliott. 2012. Effects of hemlock mortality on streams in the southern Appalachian Mountains. American Midland Naturalist 168(1) 112–131. Wehr, J.D., and R.G. Sheath (Eds.). 2003. Freshwater Algae of North America. Academic Press, Boston, MA. 918 pp. Welsh, H.H., and S. Droege. 2001. A case for using plethodontid salamanders for monitoring biodiversity and ecosystem integrity of North American forests. Conservation Biology 15(3):558–569. Whitford, L.A., and G.J. Schumacher. 1984. A Manual of Freshwater Algae. Sparks Press, Raleigh, NC. 324 pp. Willacker, J.J., Jr., W.V. Sobczak, and E.A. Colburn. 2009. Stream macroinvertebrate communities in paired hemlock and deciduous watersheds. Northeastern Naturalist 16(1):101–112. Yorks, T.E., J.C. Jenkins, D.J. Leopold, D.J. Raynal, and D.A. Orwig. 2000. Influences of Eastern Hemlock mortality on nutrient cycling. Pp. 126–133, In Sustainable Management of Hemlock Ecosystems in Eastern North America. Symposium Proceedings USDA GTR-NE (Vol. 267), 22–24 June, 1999, Durham, NH. 247 pp.