Eagle Hill Masthead

Southeastern Naturalist
    SENA Home
    Range and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Other EH Journals
    Northeastern Naturalist
    Caribbean Naturalist
    Urban Naturalist
    Eastern Paleontologist
    Eastern Biologist
    Journal of the North Atlantic

EH Natural History Home


About Southeastern Naturalist


Microstegium vimineum Invasion Changes Soil Chemistry and Microarthropod Communities in Cumberland Plateau Forests
Deborah A. McGrath and Meagan A. Binkley

Southeastern Naturalist, Volume 8, Number 1 (2009): 141–156

Full-text pdf (Accessible only to subscribers.To subscribe click here.)


Site by Bennett Web & Design Co.
2009 SOUTHEASTERN NATURALIST 8(1):141–156 Microstegium vimineum Invasion Changes Soil Chemistry and Microarthropod Communities in Cumberland Plateau Forests Deborah A. McGrath1,* and Meagan A. Binkley1 Abstract - Microstegium vimineum (Japanese Stiltgrass) is an exotic shade-tolerant C4 grass that invades open and forested habitats throughout the southeastern United States. Studies suggest that invasive plants can alter ecosystem biogeochemistry by changing soil chemistry and biota. The objective of our study was to determine if M. vimineum invasion induces soil chemical changes that alter litter microarthropod communities in acidic, nutrient-poor upland forests of the Cumberland Plateau. In a greenhouse experiment comparing forest soil in tubs seeded with M. vimineum to those left unseeded over 1 year, we found that after 6 months, soil pH under M. vimineum was significantly higher than that in the unseeded tubs. We compared A-horizon chemistry, litter nutrients, and microarthropod community diversity in 3 forested sites with and without M. vimineum. We found higher pH, phosphorus (P), and base cations, and lower aluminum (Al) in soil under dense M. vimineum growth compared to soil under surrounding uninvaded understory. M. vimineum litter was more P-rich and had a higher abundance of mites than the surrounding forest fl oor over 3 sampling periods. However, microarthropod community evenness was lower in M. vimineum litter, indicating a decrease in diversity. These results suggest that a rapid rise in soil pH and P availability following M. vimineum colonization may reduce litter microarthropod community diversity by favoring mites. Introduction The upland forests of Tennessee’s southern Cumberland Plateau are among the most diverse plant communities in the southeastern United States (Ricketts et al. 1999). The acidic sandstone-derived soils of the upland ridges support drought-resistant mixed oak-hickory forests that are increasingly undergoing conversion and fragmentation with surrounding land use change (McGrath et al. 2004), often resulting in the influx of nonnative plant opportunists. The invasion of Microstegium vimineum (Trin.) A. Camus (Japanese Stiltgrass) is of particular concern because of its traits as a “super generalist” that allow this C4 grass to invade a variety of habitats, including forest understories where it utilizes sunflecks (Cheplick 2005, Horton and Neufeld 1998, Leicht et al. 2005). Consequently, dense stands of M. vimineum are found in fields and along roadsides, as well as in forests across the region. Studies show that introduced plants can change soil processes by using or obtaining resources differently than native plants, often altering rhizosphere or litter chemistry, biomass accumulation, water relations, or microclimate 1Department of Biology, University of the South, 735 University Avenue, Sewanee, TN 37383-1000. *Corresponding author - dmcgrath@sewanee.edu. 142 Southeastern Naturalist Vol. 8, No. 1 (Ehrenfeld 2003, Vitousek 1990). By altering both habitat and substrate, soil and litter changes following plant invasion can infl uence the composition of soil biota, which in turn may affect soil food webs, litter decomposition rates, mineralization, and biogeochemical cycling within the system (Belnap and Phillips 2001, Scott et al. 2001, Yeates and Williams 2001). Researchers report higher soil pH and extractable N under M. vimineum, as well as rhizosphere microbial communities that differ from those under adjacent native plants (Ehrenfeld et al. 2001, Kourtev et al. 2002). As microbial grazers and saprophages, soil microarthropods such as mites (Acari) and collembolans (Collembola) significantly contribute to the decomposition and biogeochemical cycling of forest-floor habitats. Microarthropod populations are sensitive to disturbance and land-cover change, and a few studies demonstrate that community diversity decreases with monotypic leaf litter (Coleman et al. 2004, Hansen 2000, Migge et al. 1998). The nutrient status, physical structure, and pattern of litter decay also influences microarthropod habitat, regardless of whether the litter is comprised of one or many species (Hansen 2000). Thus, invasion-induced changes in plant communities may have a profound effect on litter microarthropod communities. While there is ample evidence that plant invasions change soil chemistry, the impact of such changes on soil microarthropods is less studied. The objective of our study was to determine if M. vimineum invasion rapidly changes soil chemistry and litter microarthropod communities, potentially altering biogeochemical processes in acidic nutrient-poor upland soils of the Cumberland Plateau. First, we conducted a greenhouse experiment to determine how M. vimineum invasion changed upland forest soil chemistry in a single growing season. We then studied mineral soil and litter chemistry with and without M. vimineum on three upland sites, each under a different forest type (mature and early successional hardwood, and a pine plantation). We hypothesized that upland soils with dense M. vimineum growth would have higher soil pH and greater cation availability than surrounding soils not colonized by M. vimineum, which might induce other changes, including differences in litter nutrients. We compared microarthropod abundance, order richness, and community diversity between M. vimineum litter and that of surrounding forest without M. vimineum on the same site. Because M. vimineum produces dense mats of monotypic litter, we predicted that M. vimineum-invaded patches would host less diverse microarthropod communities than those in the surrounding forest fl oor. Methods Species A native to East Asia, M. vimineum is recognized as a common invasive plant in 25 states from New Jersey to Texas (Barden 1987, Morrison et al. 2007). M. vimineum is a sprawling annual grass with branching stems that root from nodes, facilitating the establishment of 2009 D.A. McGrath and M.A. Binkley 143 thick, widespread infestations that are evident as dense litter mats after the plant dies. The species flowers in early fall and produces a large seed bank that remains viable for 5–7 years (Gibson et al. 2002). The small seed is dispersed easily by water and animals, facilitating invasion in disturbed locations such as floodplains, road banks, and hiking trails (Hunt and Zaremba 1992, Redman 1995). Field sites Field data were collected from 2005–2007 on upland sites near Sewanee, TN, on the southern tip of the Cumberland Plateau (35º18'N and 86º06'W). Derived from Pennsylvanian sandstone, upland soils are typically acidic, nutrient poor, and drought-prone (McGrath et al. 2004). These upland soils support oak-hickory forests dominated by Quercus prinus L. (Chestnut Oak) and Q. velutina Lam. (Black Oak), with Acer rubrum L. (Red Maple), (Nyssa sylvatica Marsh. (Black Gum), Sassafras albidum (Nutt.) Nees (Sassafras), Oxydendron arboreum (L.) DC. (Sourwood), and Vaccinium spp. (blueberry) in the middle and lower canopy. We chose 3 sites representing different forest types, all located along the same upland ridge within 10 km of each other. Originally all native oak-hickory forest, two sites had been converted to different land covers. Thus, the study included a >10-ha tract of mature (60+ years) oak-hickory forest, a 40-year old 5-ha Pinus strobus L. (White Pine) plantation, and a 5-ha agricultural homestead abandoned 15 years ago and undergoing succession (early successional forest). Although structurally different, the mature and early successional forests were comprised of species typical of the region’s upland oak-hickory complex. Greenhouse and field soil chemistry Prior to conducting our field studies, we used a greenhouse experiment to determine if the growth of M. vimineum changed forest soil chemistry in one season. Ten rectangular (36 x 30 x 14 cm) plastic basins were perforated on the bottom for drainage and filled with a 5-cm layer of sterilized sand and gravel, followed by a 15-cm layer of sieved and homogenized mineral soil collected and composited from the top 10 cm of 5 uninvaded mature oakhickory forests located within 10 km of the field sites. In June 2005, 1.0 g M. vimineum seed, harvested from wild plants during the previous season, was broadcast evenly over a 20- x 12.5-cm area (250 cm2) in 5 of the 10 basins and covered lightly with soil. Five basins were left unseeded, and all 10 basins were randomly assigned a position in the greenhouse and sprinkled daily with water using an automated system. The M. vimineum treatment germinated, densely covering the seeded area, and grew throughout the summer, ending its lifecycle after fl owering in late November. Prior to sowing the M. vimineum, a composite soil sample comprised of 5 cores extracted from the top 5 cm was collected from each basin and analyzed for pH (in a 2:1 slurry of deionized water using a pH meter), concentrations of Melich-III extractable P, K, Mg, Ca, Al, and organic matter (%), and cation exchange capacity 144 Southeastern Naturalist Vol. 8, No. 1 (barium chloride method) by A&L Analytical Laboratories (Memphis, TN). Nitrate (NO3 -), analyzed by cadmium reduction of air-dried samples, was used as an index of N availability. Soil from both the unseeded basins and beneath the M. vimineum was re-sampled and analyzed chemically twice at 6-month intervals (December 2005 and July 2006). To determine the effect of M. vimineum and time on soil chemical properties, the data were analyzed statistically using a repeated measures ANOVA with a between subject factor of treatment (seeded vs. unseeded) and a within subject factor of sampling interval (pre-seeding, and 6 and 12 months post-seeding) performed in SPSS 16.0 (SPSS, Inc., 2007). We conducted post hoc multiple comparisons using two sample t-tests to detect significant differences in mean soil properties between treatments within a sampling period In August 2006, we compared chemical properties between soils with and without dense (>90%) M. vimineum growth on 3 forested sites. In the understories of the pine, and mature and early successional oak-hickory sites, we collected 5 composite A-horizon soil samples (10 cores each, 0–5 cm depth) beneath areas covered with dense, live M. vimineum and under surrounding uninvaded forest vegetation. We realize these 5 samples do not represent a true site replication, but we were looking for initial site differences that could be explored in greater detail later. The sampled areas of M. vimineum growth were well established, varied from 4–36 m2 in size, and were located intermittently as densely colonized patches in the understory of each forest site. Uninvaded surrounding forest soil samples were collected within 3 m around the edge of each patch colonized by M. vimineum. The air-dried soil samples were sieved through a 2-mm mesh and analyzed for the same properties described for the greenhouse experiment. To compare chemical properties between soils dominated by patches of dense M. vimineum growth and surrounding uninvaded overstory, the data were analyzed using a multivariate ANOVA with factors of site (pine, mature, and early successional) and understory (invaded vs uninvaded), and their interaction (site x understory), performed in SPSS 16.0 (SPSS, Inc., 2007). Probability plots performed in SPSS confirmed that the data met the ANOVA normality assumption. Overall soil differences among the 3 sites were identified using a Bonferroni multiple comparison procedure. Litter nutrients and microarthropods In August 2005 and February 2006, we compared microarthropod abundance and diversity between oak-hickory and M. vimineum litters at the early successional forest site because it had the heaviest coverage of both litter types in close proximity. We collected seven 0.25-m2 samples of litter from patches dominated by M. vimineum and under surrounding uninvaded forest vegetation, and extracted microarthropods from 5000-cm3 subsamples over 24 hours using the Tullgren funnel method (Crossley and Blair 1991). Microarthropods dropping from the drying litter were collected in jars beneath the funnels, preserved in 70% ethanol, and examined under 2009 D.A. McGrath and M.A. Binkley 145 a dissecting (10x) microscope. The microarthropods were counted and identified by order using descriptions published by Meyer (1994). Litter dry mass was used to calculate abundance as number of microarthropods/ 100 g litter. Community diversity was calculated using the Shannon-Weiner Index: H' = -Σpi ln pi, where pi is the proportion of the total number of individuals for order i. The Shannon-Weiner index integrates both abundance and richness to provide an indicator of community evenness (Schmitz 2007). In July 2007, we repeated the study on all three forested sites (pine, early successional, and mature oak-hickory). Mean microarthropod abundance, order richness, and community diversity (H') were compared between M. vimineum and early successional oak-hickory (n = 7) litters, as well as across all three forested sites (n = 21), using a multivariate repeated measures ANOVA with a between-subject factor of litter type, a within subject factor of time (August 2005, March 2006, July 2007 sampling periods) and litter x time interaction. Post hoc t-tests identified differences between litter types within a sampling period. Following the July 2007 microarthropod collection, the litter samples were removed from the Tullgren funnels, oven dried for 24 hours, weighed, ashed in a muffl e furnace, and analyzed for nutrient concentrations using ICAP spectroscopy by A & L Analytical Laboratories (Memphis, TN). Mean elemental concentrations in litters dominated by M. vimineum and surrounding uninvaded understory vegetation (pine, mature and early successional oak-hickory) were compared statistically using a multivariate ANOVA with factors of site and litter type (M. vimineum vs uninvaded understory), and their interaction (site x litter) performed in SPSS 16.0 (SPSS, Inc., 2007). Overall differences in litter nutrients among sites were identified using post hoc Bonferroni multiple comparisons. Results Greenhouse and field soil chemistry The greenhouse seeding study revealed significant effects of M. vimineum growth, as well as time, on soil properties (Table 1). Soil pH, and extractable Al, NO3 -, P, Ca, and Mg all demonstrated significant treatment x time interactions (Figs. 1a–d). Prior to seeding M. vimineum, the greenhouse soils did not differ chemically; however, 6 and 12 months later, pH was significantly higher in soil under M. vimineum than in the unseeded basins (Fig 1a). The rise in soil pH appeared to reduce Al solubility, as indicated by lower extractable Al under M. vimineum than the unseeded soil (Fig 1b). Soil NO3 - and P concentrations in the M. vimineum treatment remained lower than those in the unseeded soils, which rose significantly after 6 and 12 months, presumably due to net N and P mineralization in the unseeded soils (Figs. 1c and d). Soil NO3 - concentrations under M. vimineum remained the same after 12 months, while P was significantly higher compared to pre-seeding concentrations, which was attributable to either lower uptake or greater solubility induced at a higher pH (Fig. 1d). Ca also had a higher mean concentration 146 Southeastern Naturalist Vol. 8, No. 1 Table 1. The effects of treatment (seeded with M. vimineum vs. bare unseeded soil), time (pre-seeding, 6 and 12 months post-seeding) and treatment x time interaction on greenhouse soil chemical properties. Data are means ± one SE; (n = 5. Significant (P < 0.05) differences between treatments within a sampling period were determined using post hoc t-tests and are indicated by *. Pre-seeding 6 months 12 months Repeated measures ANOVA P-values Soil property M. vimineum Unseeded M. vimineum Unseeded M. vimineum Unseeded Treatment Time Treatment x time pH 4.28 ± 0.02 4.29 ± 0.02 4.94 ± 0.11 4.24 ± 0.10* 4.9 ± 0.03 4.5 ± 0.1* 0.001 0.001 0.002 NO3 - 3.3 ± 0.2 2.9 ± 0.2 1.6 ± 0.3 20.8 ± 0.5* 3.4 ± 0.3 8.2 ± 2.9 0.001 0.01 0.002 P (mg/kg) 7.8 ± 0.3 7.7 ± 0.3 11.4 ± 1.1 21.4 ± 1.7* 20 ± 2 33 ± 2* 0.003 0.001 0.001 K (mg/kg) 94 ± 1 91 ± 2 62 ± 3 63 ± 4 56 ± 5 56 ± 1 0.80 0.001 0.62 Ca (mg/kg) 210 ± 8 201 ± 8 271 ± 22 273 ± 12 415 ± 28 597 ± 57* 0.05 0.002 0.02 Mg (mg/kg) 48 ± 1 45 ± 1 40 ± 5 43 ± 1 47 ± 1 39 ± 1* 0.78 0.01 0.07 Org mat (%) 6.0 ± 0.02 6.1 ± 0.02 6.0 ± 0.2 5.7 ± 0.2 4.4 ± 0.5 5.2 ± 0.5 0.81 0.07 0.73 CEC 9.0 ± 0.6 9.6 ± 0.5 7.0 ± 0.2 7.1 ± 0.2 7.4 ± 0.3 8.7 ± 0.3* 0.05 0.05 0.47 Al (mg/kg) No data: analysis not done 860 ± 18 948 ± 12* 838 ± 6 985 ± 10* 0.001 0.54 0.03 Table 2. A comparison of soil properties underlying dense patches of live M. vimineum and surrounding uninvaded forest vegetation across 3 sites (mature and early successional oak-hickory and a 40-year-old White Pine stand) on the Cumberland Plateau. Data are means ± one SE (n = 5 samples per site). Significant (P < 0.05) overall soil differences among sites were determined using a Bonferroni post hoc multiple comparison procedure and are indicated by *. Mature oak-hickory Successional oak-hickory 40-year old pine ANOVA P-values Soil property M. vimineum Uninvaded M. vimineum Uninvaded M. vimineum Uninvaded Treatment Site Treat x site pH 5.3 ± 0.1 4.5 ± 0.2 6.0 ± 0.2 5.3 ± 0.2* 5.5 ± 0.1 4.9 ± 0.1 0.001 0.001 0.76 NO3 - 4.0 ± 1.0 3.0 ± 0.1 7.4 ± 0.4 7.2 ± 0.5* 3.6 ± 0.4 4.0 ± 0.1 0.53 0.001 0.40 P (mg/kg) 19 ± 1 17 ± 1 37 ± 8 19 ± 1* 15 ± 2 12 ± 1 0.01 0.002 0.08 K (mg/kg) 71 ± 4 73 ± 6* 63 ± 2 61 ± 3 70 ± 3 56 ± 4 0.13 0.02 0.09 Ca (mg/kg) 854 ± 38 790 ± 79 1642 ± 236 946 ± 196* 1259 ± 74 892 ± 28 0.002 0.01 0.08 Mg (mg/kg) 70 ± 5 45 ± 5 56 ± 6 51 ± 3 94 ± 7 58 ± 3 0.001 0.001 0.02 CEC (cmol+/kg) 9.3 ± 0.6 10.3 ± 0.5 11.2 ± 1.1 8.8 ± 0.8 11.1 ± 0.6 9.4 ± 0.3 0.08 0.82 0.05 Al (mg/kg) 825 ± 60 1127 ± 43 535 ± 17 563 ± 25* 928 ± 76 984 ±67 0.04 0.001 0.02 Ca/Al 1.1 ± 0.1 0.70 ± 0.1 3.1 ± 0.5 1.7 ± 0.4* 1.4 ± 0.2 0.9 ± 0.04 0.003 0.001 0.15 Organuc matter (%) 5 ± 0.6 5 ± 0.3 3.6 ± 0.2 3.4 ± 0.2* 4.4 ± 0.1 4.3 ± 0.1 0.49 0.001 0.99 2009 D.A. McGrath and M.A. Binkley 147 under unseeded soil at 12 months, likely contributing to a slight rise in CEC in unseeded soils sampled at that time (Table 1). The field soil sampling showed that across the 3 forested sites, soils under dense M. vimineum had significantly higher pH, extractable P, Ca, and Mg, as well as lower extractable Al, compared to surrounding uninvaded soil on the same site without M. vimineum growth (Table 2). Subsequently, Ca and Al ratios (Ca/Al) were higher in soils under dense M. vimineum. Among the 3 sites, soil pH under M. vimineum was highest in the early successional oak-hickory forest, which corresponded with the highest P and Ca availability and Ca/Al. Uninvaded soil in.surrounding mature oak-hickory forest had the lowest pH, along with a mean Al concentration twice as high as that under the early successional site (Table 2). Neither NO3 - nor percent organic matter differed between soils with and without M. vimineum. Litter nutrients and microarthropods Across the 3 forest sites, P concentrations were significantly higher in M. vimineum litter than in surrounding forest fl oor, which corresponded to Figure 1. Greenhouse soil properties demonstrating significant (P < 0.05) treatment (seeded with M. vimineum vs. unseeded bare soil) x time (pre-seeding, and 6 and 12 months post-seeding) interactions. Extractable aluminum analyses were not performed on soils prior to seeding. For all other chemical properties, the soils showed no statistical differences prior to seeding. 148 Southeastern Naturalist Vol. 8, No. 1 Table 4. The effects of litter type (M. vimineum vs. surrounding uninvaded forest fl oor), sampling year (Aug 2005, Feb 2006, July 2007), and litter by year interaction on litter microarthropod abundance, diversity (H'), and richness were determined using a repeated measures multivariate ANOVA. In Aug 2005 and Feb 2006, litter samples were compared only on the early succesional oak-hickory site (n = 7). In July 2007, litter microarthropods were compared across the 3 sites (mature, early successional, and pine; n = 21). Significant (P < 0.05) differences between litter types within a sampling period were determined using post hoc t-tests denoted by *. Repeated measures Aug 2005 (n = 7) Feb 2006 (n = 7) July 2007 (n = 21) ANOVA P-values Uninvaded Uninvaded Uninvaded Litter type Year Treatment Soil property forest M. vimineum forest M. vimineum forest M. vimineum (treatment) (time) x time Abundance (arthropods/100 g) 547 ± 70 1144 ± 290* 172 ± 50 431 ± 183 100 ± 18 566 ± 133* 0.001 0.01 0.46 Diversity (H') 1.04 ± 0.11 0.50 ± 0.35* 0.41 ± 0.09 0.34 ± 0.06 0.61 ± 0.07 0.42 ± 0.05* 0.003 0.001 0.001 Richness (orders/sample) 8.3 ± 0.7 7.6 ± 0.6 3.2 ± 0.6 4.3 ± 0.8 3.4 ± 0.3 4.1 ± 0.4 0.55 0.003 0.29 Table 3. A comparison of elemental concentrations in M. vimineum litter and surrounding uninvaded forest fl oor across 3 sites (mature and early successional oak-hickory and a 40-year-old White Pine stand) on the Cumberland Plateau. Data are means ± one SE (n = 7 samples per forest site). Significant (P < 0.05) differences among sites across both litter types were determined using a Bonferroni post hoc multiple comparison procedure and are indicated by *. Mature oak-hickory Early successional oak-hickory 40-year old pine ANOVA P-values M. vimineum Oak-hickory M. vimineum Oak-hickory M. vimineum White Pine Treatment Element litter litter litter litter litter litter Treatment Site x site N 1.3 ± 0.09 1.6 ± 0.6 1.3 ± 0.08 1.2 ± 0.07 1.2 ± 0.05 1.1 ± 0.06 0.74 0.52 0.69 S 0.14 ± 0.01 0.12 ± 0.003 0.14 ± 0.01 0.15 ± 0.004* 0.12 ± 0.004 0.11 ± 0.002 0.14 0.0001 0.05 P 0.08 ± 0.004 0.06 ± 0.004 0.15 ± 0.01 0.11 ± 0.005* 0.08 ± 003 0.05 ± 0.003 0.0001 0.0001 0.74 K 0.25 ± 0.06 0.16 ± 0.01 0.21 ± 0.05 0.21 ± 0.01 0.18 ± 0.02 0.14 ± 0.01 0.09 0.23 0.43 Mg 0.13 ± 0.003 0.12 ± 0.004 0.08 ± 0.003 0.11 ± 0.005 0.08 ± 0.003 0.07 ± 0.005* 0.03 0.0001 0.001 Ca 0.79 ± 0.08 1.0 ± 0.05 0.70 ± 0.03 1.5 ± 0.11 0.83 ± 0.04 0.70 ± 0.02* 0.001 0.0001 0.001 Al 1489 ± 545 913 ± 138 2249 ± 333 401 ± 69 1672 ± 340 567 ± 71 0.001 0.79 0.12 2009 D.A. McGrath and M.A. Binkley 149 higher soil availability of this element (Tables 2 and 3). In contrast, Ca was lower in M. vimineum litter than in oak-hickory litter, despite higher soil extractable Ca. Litter K concentrations appeared higher in M. vimineum on 2 sites. Surprisingly, Al in M. vimineum litter was higher than in surrounding forest fl oor, even though soil Al concentrations were lower beneath the grass (Table 3). Litter type significantly affected abundance and diversity of microarthropod populations (Table 4). Microarthropod abundance was higher in the M. vimineum litter than in uninvaded understory litter from the same site across the sampling periods. Order richness did not differ between the 2 litter types, perhaps because the taxon is too inclusive (Table 4). Mite abundance was higher in M. vimineum litter and nearly 10 times greater than that of Collembola, the next most abundant order (Figs. 2a and b). However, because mites dominated M. vimineum litter, diversity (H') was higher in surrounding uninvaded litter collected in 2005 and 2007, where microarthropod abundance was spread more evenly across several orders (Table 4). Collembolan abundance exhibited no pattern with respect to litter type (Fig. 2b). Colleoptera, the third most abundant order, had equally low mean abundance in both M. vimineum and surrounding forest litter for all three sampling periods (P ≥ 0.30), ranging from a high of 11 (± 2 SE) per 100 g litter in Aug 2005 to 2 (± 0.7 SE) in the July 2007 sampling. A comparison among the sites showed that H' was higher (marginally significant) in surrounding uninvaded litter only in the 2 oak-hickory forests (Fig. 3a); however, mite abundance was greater in M. vimineum litter on all 3 sites (Fig. 3b). The repeated measures ANOVA demonstrated significant effects of sampling date on microarthropod abundance, community diversity, and order richness (Table 4). Microarthropod order richness, abundance, Figure 2. A comparison over 3 sampling periods of mean (± one SE) abundance of (a) mites and (b) collembolans between litters dominated by M. vimineum and surrounding uninvaded forest vegetation on the same sites. In Aug 2005 and March 2006, n = 7 samples of each litter type collected from the early successional forest site. In July 2007, n = 21 (3 forest sites x 7 replications). 150 Southeastern Naturalist Vol. 8, No. 1 and community diversity (H') were significantly higher in surrounding forest litter collected in August 2005 than that in the subsequent sampling periods (Table 4). The more diverse August 2005 samples contained representatives of over 10 other orders, including Diplura, Diptera, Hymenoptera, Isoptera, Lepidoptera, and Protura. Only order richness in M. vimineum litter differed significantly among the 3 sampling periods, again with the August 2005 sampling exhibiting the greatest diversity. Microarthropod abundance, order richness, and H' were similarly low in the February 2006 and July 2007 samples (Table 4), likely due to a colder, drier climate in February and a 100-year drought during the summer of 2007 (National Climate Data Center 2008). Discussion Changes in soil chemistry Although numerous studies have established the capacity of plants to acidify soil, mainly through rhizosphere production of organic acids (Kelly et al. 1998), it is less often reported that soil pH rises in response to a change in plant species. Despite a small sample size, our greenhouse study demonstrated a significant rise in soil pH that occurred within 6 months of M. vimineum germination, and confirmed that the pH change was a result of M. vimineum growth. Across the forest sites, the higher soil pH under M. vimineum was accompanied by greater availability of P and base cations Mg and Ca compared to soil under surrounding overstory vegetation. These soil changes under M. vimineum are not surprising since pH has a major role in controlling nutrient bioavailability (Kelly et al. 1998). Phosphate is more labile in soils with a pH > 5 because it does not react with Al or Fe, both of which are soluble at pH values 3–5 (Bohn et al. 2001). Base cation Figure 3. A comparison over 3 sampling periods of mean (± one SE) (a) community diversity (H') and (b) mite abundance between litters dominated by M. vimineum and surrounding uninvaded forest vegetation on 3 sites (data collected in July 2007, n = 7 for each forest site). 2009 D.A. McGrath and M.A. Binkley 151 availability can be greater in higher pH soils because the exchange complex is not dominated by hydrogen ions (H+) (Sposito 2008). A higher soil pH under M. vimineum corresponded with a significant rise in Ca/Al, due to a combination of lowered Al solubility and higher Ca concentrations. In acid soils, Ca/Al can be an indicator of forest health, with plants in soils with Ca/Al ≤ 1.0 at greater risk for Al stress, the effects of which include impaired Ca and Mg uptake, as well as inhibition of cell division and energy transfer (Borer et al. 2004, Cronon and Grigal 1995). Some Al-resistant plants exclude the uptake of this element by raising rhizosphere pH, and subsequently lowering the solubility of toxic Al (Degenhardt et al. 1998, Jansen et al. 2002). Grasses are often more sensitive to Al toxicity than other angiosperms (Jansen et al. 2002), and studies suggest that Al exposure could stimulate NO3 - uptake, raising rhizosphere pH through anti-port transport of NO3 - across the root cell membrane in exchange for a hydroxyl (OH-) anion (Degenhardt et al. 1998). Ehrenfeld et al. (2001) attributed an elevated soil pH under M. vimineum to rapid NO3 - uptake stimulated by an increase in soil nitrification rates beneath the shallow-rooted grass. We did not find an effect of M. vimineum invasion on soil NO3 - concentrations, which were equally low in dense patches of M. vimineum and surrounding overstory vegetation. However, data from the greenhouse study suggest that M. vimineum may be very effective at taking up mineralized N. After 6 months, greenhouse soil nitrate concentrations were 7-fold higher in the unseeded soil, but remained the same under M. vimineum, presumably due to uptake by the grass. In contrast, not all bioavailable P was taken up by M. vimineum, as evidenced by a rise in greenhouse soil P concentrations under the grass at 6 and 12 months. Changes in litter nutrients and microarthropods In all 3 forests, M. vimineum appeared to produce a more P-rich litter than surrounding forest fl oor, but N concentrations did not differ from overstory litter. Ehrenfeld et al. (2001) reported that M. vimineum had N-poor litter, which decomposed more slowly and immobilized N, compared to native forest litter. We observed that dead M. vimineum falls over and forms dense litter mats which may persist over multiple seasons (as evidenced by layers of hardwood litter above the M. vimineum mat), suggestive of slower decomposition relative to oak-hickory litter. In addition, M. vimineum litter had a higher Al concentration than surrounding overstory litter across the 3 sites, and studies show that this element tends to adsorb to litter exchange sites during prolonged periods of decomposition (Rustad 1994). Numerous variables are reported as factors controlling litter microarthropod community composition including nutrient availability (King and Hutchinson 1980, Lindbert and Perrson 2004), soil properties (Huhta and Ojala 2006, L’ubomir et al. 2001), plant diversity (Hansen 2000, Migge et al. 1998), litter complexity (Hansen and Coleman 1998), and disturbance (Bird et al. 2004, Reynolds et al 2007). We hypothesized that there would be greater diversity and abundance of microarthropods in 152 Southeastern Naturalist Vol. 8, No. 1 the early successional oak-hickory forest floor than in M. vimineum litter due to greater complexity of the mixed hardwood litter as a habitat. We found that M. vimineum litter had a higher abundance, but an overall lower evenness in order distribution of microarthropods than surrounding hardwood litter, due to the dominance of mites in the former. Hansen and Coleman (1998) attributed higher oribatid mite species richness and abundance to complementary resource availability provided by more complex litters with different leaf architectures, chemical characteristics, and thus differing rates of decomposition and nutrient release. After expanding our study to include the 3 forest sites, we found the same pattern of higher mite abundance in M. vimineum litter compared to the surrounding forest floor, even though oak-hickory and pine litter vary considerably with respect to litter complexity and heterogeneity. Similar to Huhta and Ojala (2006), we found no consistent effect of litter type on collembolan populations, perhaps because these microarthropods are characterized as opportunists or “r” strategists (Coleman et al. 2004), feeding upon both fungi and decomposed plant litter. Generally, mites comprise the majority of litter microarthropods, and despite high variability among sampling years, we observed densities in M. vimineum litter ≥500 mites/100 gm, at the upper end of reported values for similar sampling methods (Coleman et al. 2004). Oribatid mites are considered “K” strategists because they are long-lived fungal grazers, with both low mobility and reproductive rates that are favored by stable habitats (Coleman et al. 2004). Hansen (2000) hypothesized that oak litter, which breaks down slowly initially but decomposes rapidly later in the season, produces fl ushes and troughs of microbial growth, rendering it an unstable habitat for mites. In contrast, Ehrenfeld et al. (2001) found that dead M. vimineum blades broke down quickly, but the remaining grass culms decomposed slowly. Therefore, we believe that the slowly decomposing M. vimineum litter provides a more sustained supply of P for microbial growth than surrounding forest fl oor, thereby creating a more stable habitat for mites. Studies have shown an increase in microarthropod abundance in response to fertilization, which raises both litter nutrient concentrations and plant productivity (Bird et al. 2004, Lindbert and Persson 2003). King and Hutchinson (1980) demonstrated a 4-fold increase in grassland mite densities following superphosphate application, with litter P concentration providing a better predictor of microarthropod abundance than N. Thus, higher soil and litter P availability, induced by a rise in pH following M. vimineum invasion may favor litter mite proliferation, perhaps at the expense of overall microarthropod community evenness. Higher soil Ca availability under M. vimineum may also contribute to their abundance, because oribatid mites are known to sequester Ca (metabolized from fungi) in exoskeletons (Coleman et al. 2004). In addition, the dense M. vimineum litter mat may promote moisture retention (Ehrenfeld et al. 2001) or moderate temperature extremes, either of which could increase habitat stability. 2009 D.A. McGrath and M.A. Binkley 153 Potential effects on biogeochemical cycling The most consistent effect we found was that M. vimineum invasion elevated soil pH and P availability, and lowered litter microarthropod community diversity by significantly increasing the abundance of mites. It is unclear how higher mite abundance in response to M. vimineum invasion would affect biogeochemical cycling. Mites contribute to litter decomposition by mobilizing nutrients and fragmenting litter as they graze upon bacteria and fungi, but their direct effect on litter mass loss is minimal (Beare et al. 1992). However, a higher mite abundance and lower microarthropod order evenness would likely change soil food-web dynamics, as mites are prey for many larger organisms such as beetles, ants, and salamanders (Coleman et al. 2004). The broader question of whether a change in soil pH is enough to alter soil biogeochemistry deserves further examination. A comparison among the 3 forest sites yields limited inferences without replication of each forest type; however, it is likely that the effect of M. vimineum invasion on soil chemistry depends upon the initial soil characteristics and the plant community overlying it. For example, only soils under mature oak-hickory forest had Ca/Al < 1, which rose to > 1 under adjacent patches of M. vimineum (theoretically lowering the risk of Al-toxicity) due to a pH-induced decrease in Al solubility. Moreover, the early successional oak-hickory site, which had the highest soil pH among the sites under both M. vimineum and surrounding overstory also had significantly greater soil P and Ca concentration, the highest mite abundance, and the lowest Al concentration. This observation, although requiring further study, suggests that there may be a pH threshold at which a rise in soil pH induced by M. vimineum invasion triggers large changes in soil chemistry and biota. Another interesting question is whether a rise in pH and other soil changes following M. vimineum invasion facilitates changes in plant community composition by rendering soil more favorable for other invaders, such as Lespedeza cuneata (Dum. -Cours.) G. Don (Chinese Lespedeza), or Al-sensitive native species, such as Acer saccharum Marshall (Sugar Maple). In summary, while we found clear and rapid effects of M. vimineum invasion on soil pH, ecosystem P availability, and litter mite abundance, how these changes alter biogeochemical cycling in upland forests of the Cumberland Plateau merits further study. Acknowledgments This study was funded by grants from the DuPont and Lilly Foundations. We thank Nick Hollingshead of the University of the South’s Landscape Analysis Lab for statistical consult, Kara Allen for her assistance in data collection and analysis, and Natasha Cowie for her work establishing the greenhouse experiment. We are grateful for the helpful comments made by Cynthia D. Huebner and three anonymous reviewers. 154 Southeastern Naturalist Vol. 8, No. 1 Literature Cited Barden, L.S. 1987. Invasion of Microstegium vimineum (Poaceae), an exotic, annual, shade-tolerant, C4 grass, into a North Carolina fl oodplain. American Midland Naturalist 118(1):40–45. Beare, M.H., R.W. Parmelee, P.F. Hendrix, and W. Cheng. 1992. Microbial and faunal interactions and effects on litter nitrogen and decomposition in agroecosystems. Ecological Monographs 62(4):569–591. Belnap, J., and S.L. Phillips. 2001. Soil biota in an ungrazed grassland: Response to annual grass (Bromus tectorum) invasion. Ecological Applications 11:1261– 1275. Bird, S.B., R.N. Coulson, and R.F. Fisher. 2004. Changes in soil and litter arthropod abundance following tree harvesting and site preparation in a Loblolly Pine (Pinus taeda L.) plantation. Forest Ecology and Management 202(1–3):195–208. Bohn, H.L., B.L. McNeal, and G.A. O’Connor. 2001. Acid Soils. Pp. 48–66, In Soil Chemistry. Third Edition. John Wiley and Sons, New York, NY. 303 pp. Borer, C.H., P.G. Schabert, D.H. DeHayes, and G.J. Hawley. 2004. Accretion, partitioning, and sequestration of calcium and aluminum in Red Spruce foliate: Implications for tree health. Tree Physiology 24:929–939. Cheplick, G.P. 2005. Biomass partitioning and reproductive allocation in the invasive, cleistogamous grass Microstegium vimineum: Infl uence of the light environment. Journal of the Torrey Botanical Society 132(2):214–224. Coleman, D.C., D.A. Crossley, and P.F. Hendrix. 2004. Microarthropods. Pp. 98–124, In Fundamentals of Soil Ecology. Second Edition. Elselvier Academic Press, Burlington, MA. 375 pp. Cronan, C.S., and D.F. Grigal. 1995. Use of Calcium/Aluminum ratios as indicators of stress in forest ecosystems. Journal of Environmental Quality 24:209–226. Crossley, D.A., Jr., and J.M. Blair. 1991. A high efficiency, “low-technology” tullgren-type extractor for soil microarthropods. Agriculture, Ecosystems and Environment 34:187–192. Degenhardt, J., P.B. Larsen, S.H. Howell, and L.V. Kochian. 1998. Aluminum resistance in the Arabidopsis mutant alr-104 is caused by an aluminum-induced increase in rhizosphere pH. Plant Physiology 117:19–27. Ehrenfeld, J.G. 2003. Effects of exotic plant invasions on soil nutrient processes. Ecosystems 6:503–523. Ehrenfeld, J.G., P. Kourtev, and W. Huang. 2001. Changes in soil functions following invasions of exotic understory plants in deciduous forests. Ecological Applications 11(5):1287–1300. Gibson, D.J., G.Spyreas, and J. Benedict. 2002. Life history of Microstegium vimineum (Poaceae), an invasive grass in southern Illinois. Journal of the Torrey Botanical Society 129(3):207–219. Hansen, R.A. 2000. Effects of habitat complexity and composition on a diverse litter microarthropod assemblage. Ecology 8(4):1120–1132. Hansen, R.A., and D.C. Coleman. 1998. Litter complexity and composition are determinants of the diversity and species composition of oribatid mites (Acari: Oribatida) in litterbags. Applied Soil Ecology 9:17–23. Horton, J.L., and H.S. Neufeld. 1998. Photosynthetic responses of Microstegium vimineum (Trin.) A. Camus, a shade-tolerant, C4 grass, to variable light environments. Oecologia. 114:11–19. 2009 D.A. McGrath and M.A. Binkley 155 Huhta, V., and R. Ojala. 2006. Collembolan communities in deciduous forests of different origin in Finland. Applied Soil Ecology 31:83–90. Hunt, D.M., and R.E. Zaremba. 1992. The northeastward spread of Microstegium vimineum (Poaceae) into New York and adjacent states. Rhodora 94:167–170. Jansen, S., M.R. Broadley, E. Robbrecht, and E. Smets. 2002. Aluminum hyperaccumulation in Angiosperms: A review of its phylogenetic significance. The Botanical Review 68(2):235–269. Kelly, E.F., O.A. Chadwick, and T.E. Hilinski. 1998. The effect of plants on mineral weathering. Biogeochemistry 42:21–53. King, K.L., and K.J. Hutchinson. 1980. Effects of superphosphate and stocking intensity on grassland microarthropods. The Journal of Applied Ecology 17(3): 581–591. Kourtev, P.S., J.G. Ehrenfeld, and M. Haggbloom. 2002. Exotic plant species alter the microbial community structure and function in the soil. Ecology 83:3152–3166. Leicht, S.A., J.A. Silander, Jr., and K. Greenwood. 2005. Assessing the competitive ability of Japanese Stilt Grass, Microstegium vimineum (Trin.) A. Camus. Journal of the Torrey Botanical Society 132 (4):573–580. Lindbert, N., and T. Persson. 2003. Effects of long-term nutrient fertilization and irrigation on the microarthropod community in a boreal Norway Spruce stand. Forest Ecology and Management 188(1–3):125–135. L’ubomir, K., P. Luptacik, D. Miklisova, and R. Mati. 2001. Soil oribatida and collembola communities across a land depression in an arable field. European Journal of Soil Biology 37(4):285–289. McGrath, D.A., J.P. Evans, C.K. Smith, D.G. Haskell, N.W. Pelkey, R.R. Gottfried, C.D. Brockett, M.D. Lane, and E.D. Williams. 2004. Mapping land-use change and monitoring the impacts of hardwood-to-pine conversion on the southern Cumberland Plateau in Tennessee. Earth Interactions 8(9):1–23. Meyer, J.R. 1994. Kwik-Key to Soil-Dwelling Invertebrates. Department of Entomology. North Carolina State University, Vision Press, Raleigh, NC. 43 pp. Migge, S., M. Maraun, S. Scheu, and M. Schaefer. 1998. The oribatid mite community (Acarina) of pure and mixed stands of beech (Fagus sylvatica) and spruce (Picea abies) of different age. Applied Soil Ecology 9(1–3):115–121. Morrison, J.A., H.A. Lubchansky, K.E. Mauck, K-M. McCartney, and B. Dunn,. 2007. Ecological comparison of two co-invasive species in eastern deciduous forests: Alliaria petiolata and Microstegium vimineum. Journal of the Torrey Botanical Society 134(1):1–17. National Climate Data Center (NCDC). 2008. Climate of 2007—Tennessee Moist. Status. Available online at http://lwf.ncdc.noaa.gov/oa/climate/research/prelim/ drought/st040dv00pcp.html. Accessed 06 February 2008. Redman, D.E. 1995. Distribution and habitat types for Nepal Microstegium [Microstegium vimineum (Trin.) Camus] in Maryland and the District of Columbia. Castanea 60(3):270–275. Reynolds, B.C., J. Hamel, J. Isbanloly, L. Klausman, and K.K. Moorhead. 2007. From forest to fen: Microarthropod abundance and litter decomposition in a southern Appalachian fl oodplain/fen complex. Pedobiologia 51(4):273–280. Ricketts, T.H., E. Dinerstein, D. Olson, C.J. Loucks, W. Eichbaum, D. Della Sala, K. Kavanagh, P. Hedao, P. Hurley, K. Carney, R. Abell, and S. Walters. 1999. Terrestrial Ecoregions of North America: A Conservation Assessment. Island Press, Washington, DC. 491 pp. 156 Southeastern Naturalist Vol. 8, No. 1 Rustad, L. 1994. Elemental dynamics along a decay continuum in Red Spruce ecosystem in Maine, USA. Ecology 75(4):867–879. Schmitz, O.J. 2007. Biodiversity and habitat fragmentation. Pp. 80–81, In Ecology and Ecosystem Conservation. Island Press, Washington, DC. 159 pp. Scott, N.A., S. Saggar, and P.D. McIntosh. 2001. Biogeochemical impact of Hieracium invasion in New Zealand’s grazed tussock grasslands: Sustainability implications. Ecological Applications 11(5):1311–1322. Sposito, G. 2008. Exchangeable ions. Pp. 219–239, In The Chemistry of Soils, Second Edition. Oxford University Press, New York, NY. 321 pp. SPSS, Inc. 2007. SPSS 16.0 Statistical Analysis for Mac. Chicago, IL. Vitousek, P.M. 1990. Biological invasions and ecosystem processes: Towards an integration of population biology and ecosystem studies. Oikos 57:7–13 Yeates, G.W., and P.A. Williams. 2001. Infl uences of three invasive weeds and site factors on soil microfauna in New Zealand. Pedobiologia 45:367–383.