American chestnut soil carbon and nitrogen dynamics: Implications for ecosystem response following restoration

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

Contents lists available at ScienceDirect

Pedobiologia - Journal of Soil Ecology journal homepage: www.elsevier.com/locate/pedobi

American chestnut soil carbon and nitrogen dynamics: Implications for ecosystem response following restoration Geoffrey W. Schwanera, Charlene N. Kellyb, a b

T

⁎

National Ecological Observation Network, Aquatics, NEON Domain 07, 154 Fairbanks Rd, Oak Ridge, TN, 37830, United States West Virginia University, 1145 Evansdale Drive, 322 Percival Hall, Morgantown, WV, 26506, United States

A R T I C LE I N FO

A B S T R A C T

Keywords: American chestnut Tree species Soil carbon Nitrogen mineralization Oxidizable carbon

American chestnut (Castenea dentata), once dominant throughout the eastern deciduous forest of North America, was extirpated from its native range by chestnut blight fungus. Through development of blight-resistant trees, the reintroduction of chestnut is likely, though little is known about the biogeochemistry of forests influenced by chestnut. We performed a one-year laboratory incubation experiment with soil and litter from 10-year-old monoculture plantings of pure American chestnut, black cherry, and northern red oak, in addition to a fieldbased 13C isotopic analysis of soil C. Parameters included litter decomposition, C respiration, N leaching, soil oxidizable C, extracellular enzyme activity related to nutrient acquisition, and litter chemistry. Results indicate that chestnut litter decayed more rapidly than that of oak or cherry (19.0%, 10.8%, 14.1% mass loss in chestnut, oak, and cherry litter, respectively). Chestnut had lower N leaching rates than soils beneath oak or cherry (7.8, 11.5, and 12.0 mg N kg−1 in chestnut, cherry, and oak soils, respectively), greater dissolved organic C (DOC) in leachate than soils influenced by oak (32.2 and 26.4 mg kg−1 in chestnut and oak soil, respectively). No differences in soil respiration or total soil C by species were detected. We conclude that surface soils influenced by chestnut have large inputs of C through rapid litter decomposition and low inorganic N availability, indicating potential for accumulation of C in surface soil over the long-term.

1. Introduction The once-dominant American chestnut tree (Castenea dentata) was essentially extirpated from the US eastern deciduous forest as a result of the infestation by the introduced chestnut-blight fungus (Cryphonectria parasitica), beginning in the early 1900′s. The historic range of the American chestnut once covered more than 800,000 km2 and could make up more than half of the basal area of a forest stand (Braun, 1950). The loss of the American chestnut is thought to be one of the largest disturbance events in this ecosystem in the post-glacial era. Chestnut trees have been naturally replaced in forest stands by various species, but commonly by hickory (Carya spp.), yellow poplar (Liriodendron tulipifera), and oaks (Quercus spp.) (Nelson, 1955). Through a combination of backcross-breeding (Hebard, 2006) and more recent transgenic approaches (Zhang et al., 2013), the reintroduction of chestnut throughout the eastern deciduous forest is considered “imminent” (Jacobs, 2007) and engenders great public support. In the event of a successful reintroduction of the American chestnut, significant ecosystem changes in C and nutrient cycling and ecosystem water availability are expected (Ellison et al., 2005), but have yet to be

⁎

understood or quantified at a landscape level. Significant work has been done involving the silvicultural prescriptions and habitat preferences of chestnut (Jacobs and Severeid, 2004; McCament and McCarthy, 2005; Jacobs, 2007; Rhoades et al., 2009), indicating that chestnut is a generalist, intermediate shade-tolerant species, adapted to a relatively broad range of site conditions. It is reported that chestnut is highly competitive and fast-growing during the juvenile phase and it is readily capable of out-competing and replacing other native species such as oak (Quercus) and hickory (Carya) species for canopy dominance. McEwan et al. (2006) documented rapid growth of regenerating chestnut trees in Wisconsin, USA following a logging event greater than co-occurring species, indicating that if chestnut is successfully restored, it will be readily maintained in managed forests. The widespread and rapid growth, combined with chestnut’s importance for wildlife, caused it to be labeled a “foundation species”, influencing community structure and ecosystem processes wherever the species occurred (Youngs, 2000; Ellison et al., 2005). Belowground nutrient cycling can be influenced by differences in chemical composition of leaf litter (McClaugherty et al., 1985), root litter and exudates (Lin et al., 2017; Sun et al., 2018), and mycorrhizal

Corresponding author. E-mail addresses: [email protected] (G.W. Schwaner), [email protected] (C.N. Kelly).

https://doi.org/10.1016/j.pedobi.2019.05.003 Received 6 February 2019; Received in revised form 17 May 2019; Accepted 21 May 2019 0031-4056/ © 2019 Elsevier GmbH. All rights reserved.

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

plantings of pure American chestnut (ECM-fungal associated), black cherry (AM-fungal associated) and northern red oak (ECM-fungal associated), in addition to a field-based 13C isotopic analysis of soil C derived from each tree species. Fitting within the ECM-AM framework of nutrient cycling, we tested four main hypotheses. First, we hypothesize that American chestnut litter will have greater tannin and lignin content than AM-associated trees, represented in this study by black cherry. These chemical characteristics will lead to slower leaf litter decomposition. Second, high tannin release from leaf litter and slower decomposition will lead to low N availability, and thus, low N leaching. Third, N limitation in soil will lead to high C leaching and low C mineralization rates. Lastly, nutrient limitation will lead to increased soil C storage beneath chestnut in surface soil, especially relative to black cherry. We also explored nutrient-acquiring extracellular enzyme activity to investigate potential interactions with nutrient limitations and/or tannin sorption and the effect on resultant decomposition and mineralization patterns from the soil and litter incubation.

association of different tree species (Yin et al., 2014; Averill et al., 2014). Chemical composition of the leaf and root litter influence stable soil C, N availability, and microbial community and activity, which, in turn, affects forest productivity (Peterjohn et al., 1999; Magill et al., 2000; Phillips et al., 2013). Controls on nutrient cycling, particularly N, play a large role in the C budget of ecosystems (Resh et al., 2002; Phillips et al., 2013). Because of this, different tree species can be associated with divergent rates of ecosystem N cycling and loss (Lovett et al., 2002; Christopher et al., 2006; Lin et al., 2017) and storage of C, both above- and below-ground (Finzi et al., 1998; Schulp et al., 2008). The differences in C stocks can range from minimal to up to an 80-ton ha−1 difference in total C in soil and biomass between hardwood species (Jandl et al., 2007). For example, at the Canaan Mountain Plateau in Connecticut, US, red oak (Q. rubra) stands were reported to contain 9.4 kg C m-2 within the surface 15 cm of mineral soil, signifying a 16% greater C storage compared to sugar maple (Acer saccharum) stands (Finzi et al., 1998). Thus, soil nutrient cycling and C storage are susceptible to changes in dominant vegetation that may occur as a result of climate change or disturbances such as harvest, management influences, or loss of a dominant species by disease (Finzi et al., 1998). In a larger context, it is unclear how shifts in dominant tree species may impact, or have historically impacted, long-term soil organic matter (SOM) storage. Important shifts in forest biogeochemical processes likely occurred throughout the historic range of chestnut following its decline. This may have influenced standing pools of C and N within both soil and standing biomass, although the magnitude and direction of changes are unknown. However, it is known that American chestnut trees are associated with ectomycorrhizal (ECM) fungi (Jacobs et al., 2013), and trees with ECM associations generally have poorer litter quality (e.g. high C:N, high lignin:N) and slower below-ground nutrient cycling when compared to trees with arbuscular mycorrhizal (AM) associations. ECM associations generally result in greater C storage, particularly in forest floor and surface soils, when compared to AM species (Vesterdal et al., 2013; Averill et al., 2014). However, AM associations have been related to accumulation of microbially-processed C within clay mineral aggregates in soils at greater depth (Craig et al., 2018). Despite this, very little is known regarding the influence that American chestnut has on ecosystem C and N cycling relative to commonly co-occurring tree species or species that chestnut may replace, especially in below-ground processes (Rhoades, 2007). Previous studies show inconclusive evidence of chestnut litter quality; some studies report chestnut litter has lower C:N ratios than other ECM species, such as the oaks and hickories that have functionally replaced them (Ellison et al., 2005). Chestnut litter also reportedly contains greater concentrations of N, phosphorus, and potassium compared to mixed hardwood litter, particularly when growing on finetextured soil (Rhoades, 2007). In this same study, chestnut influence resulted in a lower N mineralization, but only on the coarser-textured soil. Conversely, other studies report no differences in C:N ratio between chestnut and other ECM tree litter (Rosenberg, 2010). Given that wood produced by chestnut is relatively high in tannin content, as evidenced by the historical use of chestnut wood in the leather tanning industry (Wang et al., 2013a), chestnut litter may also contain high concentrations of tannins. Higher concentrations of tannins may impact soil nutrient cycling, as tannins are known to inhibit digestive enzymes and precipitate proteins, which may slow N mineralization (Hagerman and Butler, 1981; Lovett et al., 2004) and influence soil C dynamics. Therefore, it is important to accurately understand the quality of chestnut litter relative to contemporary dominant tree species. The objective of this study was to examine how American chestnut may differ from other co-occurring hardwood species in below-ground nutrient cycling processes, including N mineralization, litter chemistry and decomposition, soil respiration, active C pools, and extracellular enzyme activity. We performed a one-year laboratory soil incubation experiment with soil and litter collected from 10-year-old monoculture

2. Methods 2.1. Study site Soil and litter samples were collected at Purdue University’s Martell Research Forest in West Lafayette, Indiana, USA (40º 26′ 42″ N, 87º 01′ 47″ W). This 2.4-ha plantation of pure, non-hybrid American chestnut, along with two other commonly co-occurring species (northern red oak, Quercus rubra, and black cherry, Prunus serotina) were planted in 2007 to study chestnut growth (Gauthier et al., 2013). The plantation is comprised of seven species compositions; pure stands of each species, two-way mixes of each species, and a three-way mix of each species. This study utilized plots of the pure monoculture stands of the three tree species. The tree spacing regimes included in this study consisted of the 1 x 1 m (10,000 stems ha−1, 5 x 5 m plot size) and 2 x 2 m (2500 stems ha−1, 10 x 10 m plot size) spacing, resulting in eight plots comprising an experimental block. This is replicated three times at the site (3 species, 2 densities, 3 replications; n = 6; N = 18 plots). At the time of soil collection, trees were 10 years old, with mean tree diameter at breast height (DBH) was 7.10, 8.23, and 8.36 cm for cherry, chestnut, and oak, respectively across all plots. Mean tree height was 4.46, 6.49, and 6.53 m for cherry, chestnut, and oak. Plots were typically under closed canopy. Soils are the Rockfield series, which are mildly acidic to nearly neutral, mainly silt loams, with a clay content of approximately 20–32 % (Soil Survey Staff, 2016), are typically moderately productive, deep, and formed from silty outwash and loamy till. Soil profiles were similar at all three blocks, showing a 2–5 cm agriculturally disturbed Ap horizon, and weak Bt-horizon development, and no significant surface organic O-horizon development. There are no measurable differences in current total soil C and N content or soil pH between species and spacing treatments, as analyzed by horizon (Ap, Bt1, Bt2) in 2017. Mean soil C across the plots was 5866 kg ha−1 and soil pH in the Ap horizon was 5.76. Prior to 2006, plots were maintained in corn production. From 1981–2010, mean annual temperature was 10.4 °C, and mean annual precipitation was 970 mm (National Climatic Data Center 2018). 2.2. Incubation core construction and sampling PVC cores used for the incubation study (20 cm height x9.28 cm diameter), were fitted with caps at the bottom, and a port opening for leachate collection. Poly-fil® material (approx. 2.5 cm) was placed at the bottom of each core to prevent soil from clogging fittings or exiting cores. Approximately 1 kg of soil was collected from the surface 0–10 cm from 10 randomized locations within one plot and composited. Soils were air-dried and sieved to 2-mm and a subsample from each plot was used to determine soil moisture by weighing before- and after oven25

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

methylumbelliferyl β-glucopyranoside as substrates, respectively. Activity was determined through fluorescence at 265 nm excitation and 460 nm emission. PPO and PER used 25 mM 3,4-L- dihydroxyphenylalanine (L-DOPA) mixed with 50 mM ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA) as a substrate and absorbance read at 460 nm. The ratio of BG (C activity) to NAG (N activity) was calculated and used to determine relative proportion of resource allocation for acquiring C and N (Knelman et al., 2017). Soil oxidizable C was measured as permanganate oxidizable C (POXC; Culman et al., 2012). POXC has been used as a proxy for total biologically active C, including particulate organic C, microbial biomass C, and SOC, and is a useful way to determine differences in potential C sequestration among different soils or treatments (Culman et al., 2012). Subsamples were air-dried and added to 0.2 M KmnO4. Solutions were shaken for 2 min at 120 rpm and settled for 10 min. Supernatant was pipetted onto a clear 96-well plate, along with permanganate standards and DI blanks, and absorbance read at 550 nm. Acid-insoluble and polyphenolic tannin content were determined from ground litter. Acid-insoluble and recalcitrant compound content of the litter was determined using a slightly modified version of the Klasson acid digestion method adapted for use with litter (Ibáñez and Bauer, 2014). Two replicates per sample of 300 mg of ground, ovendried litter were digested in 8 ml of cold (12–15 °C) 72% H2SO4 for two hours at room temperature, diluted to 3% with DI water and autoclaved at 121 °C for one hour. Samples were stored at 4 °C overnight and filtered through a ceramic crucible to separate soluble lignin and insoluble compound fractions. A 10-ml aliquot of the filtrate was saved to measure the soluble lignin fraction. Crucibles were washed free of acid, oven-dried for 16 h at 105 °C, and allowed to cool in a desiccator. The soluble fraction was measured at 280 and 215 nm on a plate reader and was calculated by the following formula: S = ((4.53 * A215) – A280))/ 300, where S is the concentration of soluble lignin, A215 is absorbance at 215 nm, and A280 is absorbance at 280 nm (Moreira-Vilar et al., 2014). Total recalcitrant compounds were measured as the sum of insoluble material that did not pass through the crucible and soluble lignin in the supernatant. Proportion of soluble lignin and insoluble compounds lost during incubation was calculated by subtracting concentration of each sample from the mean value of composited non-decayed litter by spacing and species treatment. Litter tannin content was determined using the proanthocyanidin (PA) methanol-acid assay (Preston et al., 1997; Preston, 1999), slightly modified for use with ground litter (Lorenz et al., 2000). 5% concentrated HCl in n-butanol, with a total water content of 5% v/v and 200 mg l−1 of FeSO4*7H2O was prepared fresh daily. PA standard was prepared using pycnogenol derived from pine bark (Chen and Sang, 2014) diluted with methanol. 25 mg ground litter were weighed into 50 ml centrifuge tubes with 20 ml of acetone (70% by volume), and shaken for 1.5 h, and centrifuged (7000 rpm for 20 min), and supernatant poured into 50 ml flasks. This process was repeated, and supernatants combined and brought to 50 ml total volume per sample using acetone solution. 2 ml of supernatant was transferred into tubes, and air-dried inside a fume hood, along with the pellets of remaining insoluble residue within centrifuge tubes. Upon drying, 5 ml of FeSO4 solution was added, vortexed, and placed in a 95 °C water bath for 1.5 h and allowed to cool. 300 μl of solution was transferred to 96-well plates, with methanol blanks and tannin standards, and absorbance recorded at 550 nm. Total tannin content of litter was the sum of residual tannins from the residue samples and the extractable tannins from supernatant samples. Tannin degradation during the incubation period was calculated by subtracting tannin content of each sample at T = 1 y from the mean value of composited, non-decayed litter by spacing and species treatment from T = 0.

drying at 105 ºC. Soil from each plot was separated into 3 laboratory replicates to the equivalent of 200 g dry weight of soil (N = 54 incubation units). Soil of each replicate was mixed with 160 g acid-washed sand (to maintain hydraulic conductivity for leachate sampling) and placed in cores, allowing leachate to be sampled without disturbing the soil (Giardina et al., 2001). 3.5 g dry-weight litter material from each plot (collected from plastic litter traps in October 2016 and airdried) was added whole atop of the soil/sand mixture of the corresponding tree species. 3.5 g of litter represents the average amount of leaf fall collected from the litter traps, corrected to area of the incubation cores. Litter was separated from soil using 0.3 mm fiberglass screen to allow litter contact with soil, but not mix with, to allow for collection at the conclusion of the one-year incubation. 2.3. Repeated sample collection and measurements Weekly, cores were watered to field capacity with deionized (DI) water. Field capacity was determined by adding water until it began dripping from each core. Cores were weighed when dripping ceased, and field capacity was maintained by adding water to desired weight in subsequent weeks of the experiment. Beginning in December 2016, incubation cores were measured bimonthly for soil respiration (LI−COR 8100 IRGA chamber, LI−COR Biosciences, Lincoln, NE). Cores were watered to field capacity the day before respiration measurements to ensure consistent soil moisture conditions. CO2 flux reported by the Li−COR was converted into total C respired per unit area. Monthly, 75 ml of Hoagland’s rainwater nutrient solution, modified to remove N, was added to cores after watering to field capacity, with the exception of T = 0, where DI water was used in place of Hoagland’s. By removing N from the rainwater solution, inorganic N in leachate could be measured to approximate N mineralization rates within the soil due to microbial activity. Cores were allowed to drain by gravity overnight. Typically, 75–80 % of the original volume of solution was recovered. Leachate samples were split into two 20 ml subsamples, one analyzed for DOC, and one for NO3 and NH4 N. Samples were stored frozen until analyzed. DOC was analyzed using a Shimadzu TOC analyzer (Shimadzu Scientific Instruments, Kyoto, Japan). Nitrate (NO3) and ammonia (NH4) were analyzed colorimetrically (Synergy HTX plate reader, Biotek, Winooski, VT). NO3 samples were oxidized using vanadium chloride and read at 549 nm, and NH4 used sodium salicylate and bleach, and were read at 650 nm (DeForest, 2013). Samples were assayed with four replicates per sample in clear 96-well plates, along with NH4 and NO3 standard curve samples. DOC and N concentrations were converted into mass kg−1 soil using volume of leachate collected and mass of soil within cores. 2.4. T = 1-year measurements Upon completion of the incubation, litter was dried at 65 ºC for 96 h and weighed to measure mass loss and decomposition rates. Soils were removed and stored at 4 °C until further analysis. A subsample of soil was measured for enzyme activity using methods adapted from Sinsabaugh (2008), including acid phosphatase (AP – breaks down phosphate esthers), β-N-acetylglucosaminidase (NAG – degrades organic N compounds), β-glucosidase (BG – converts cellulose into simple sugars), and phenol oxidase (PPO) and peroxidase (PER – degrades polyphenolics such as lignin and tannins) (Sinsabaugh et al., 2002; Rosenberg, 2010). AP, NAG and BG activity were determined through fluorescence, while PPO and PER were determined colorimetrically. Soil was mixed in a slurry with sodium acetate buffer solution (pH 5.0) and pipetted into opaque 96-well plate for fluorometric enzymes and clear plates for colorimetric enzymes. Plates were arrayed with standards and controls. AP, NAG and BG used 100 μM 4-methylumbelliferone (MUB) as a standard, and 4-methylumbelliferyl phosphate, 4methylumbelliferyl N-acetyl-β-Dglucosaminide and 4-

2.5. Isotopic sampling and analyses For the isotopic study, we utilized the change in land use from C4 26

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

Fig. 1. Mean percent mass loss of litter at T = 1-year (top left), lignin (top right) and tannin (bottom left) content, and C:N ratio (bottom right) of litter material at initiation and after one year from black cherry, chestnut, and red oak incubation cores. Bars with different letters within each time point are significantly different by Tukey’s HSD test. Error bars represent +/- 1 SE from the mean. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

Ap horizon soils was also determined via combustion in a CNS elemental analyzer (NA 1500 Series 2 Carlo Erba Instruments).

agriculture to C3 forest plantations to quantify species-specific differences of new inputs of tree-derived C3 relative to older soil C derived from C4 vegetation (corn) and assess differences in C accumulation over the age of the plantation. Soil was collected from the upper three mineral horizons of each of the 18 study plots (Ap, Bt1, Bt2). Samples were composited from three cores collected from random points in each plot. Foliar samples were collected using a pole saw from three randomly selected interior trees per plot and three leaves were collected from each tree canopy exposed to full or partial sunlight. In addition, samples from each horizon and corn foliar samples were collected from outside the study area from an adjacent field, to be used as the second input with collected foliage. As no C3 photosynthesis was present at this exterior corn site, it could be used to approximate the δ13C of corn inputs to the study area prior to the planting of the plantation, though with an additional 10 y of C4 influence. This biased control is able to provide a C4 index, and the differences measured in C3 indicate differences by tree species in the plots. Approximately 150 mg of sieved soil from each sample were weighed into 2 ml centrifuge tubes. Each sample was treated with 1 M HCl, vortexed for two minutes, and allowed to sit overnight to allow carbonate to diffuse out. HCl was decanted from the centrifuge tubes, and deionized (DI) water was added. Samples were centrifuged for 3 min at 3000 rpm and allowed to settle before decanting the DI water. This process was repeated until the pH of the decanted DI water was equal to DI water before added to soil. Foliage and corn residue were ground and were weighed into sample tins and analyzed for the ratio of 12 C and 13C using a Thermo Finnigan Delta gas isotope ratio mass spectrometer connected to a High-Temperature Conversion Element Analyzer and GC IsoLink (Thermo Fisher Scientific). For each species, soil, litter, and corn residue δ13C were used to evaluate the proportion of C3- and C4-derived soil C. A sub-sample of foliage was used to measure bulk C, N, and C:N ratio as described above. To determine total soil organic C (SOCT), bulk density was determined in the Ap horizon using a slide hammer to collect intact soil cores, which were dried and weighed to determine mass and volume. Carbon concentration from the

2.6. Data analysis Data were compiled and statistically analyzed using SAS-JMP software (v. 13.0). Multiple two-way ANOVA tests were used to test for significant differences in mean values, with tree species and spacing as the model effects. Because tree spacing was not a significant factor for many of the parameters measured (p > 0.05), including N and DOC in leachate, litter mass loss, POXC, litter C:N, litter lignin, AP and NAG activity, and the BG:NAG ratio, we do not include tree spacing in further discussion of the results. We note however, that tree spacing was a significant factor for measures of CO2 respiration, litter tannin at T = 1 y, and PPO and PER enzyme activity, and we report the overall effect in the results for these parameters. Further discussion of the effect of tree spacing is outside the scope of this report. Residuals were checked for normality using the Shapiro-Wilk test and transformations applied as needed. Tukey-Kramer HSD tests were used in post-hoc analysis for comparisons between significant model effects. Pairwise correlations were performed on litter mass loss and N mineralization and potential explanatory variables. The δ13C input values for the litter and soil from each plot were used in a simple 2-point mixing equation to calculate the relative abundance of C3 and C4 derived SOM in each sample. Percent of total SOM derived from C4 was calculated first (Volkoff and Cerri, 1987): % C4 = [(δ13Csample δ13C3)/ δ13C4 – δ13C3)] * 100 % C3 = 100 – % C4 Total SOC (SOCT) in kg C m−2 was calculated based on the soil bulk density and C concentration for the Ap horizon sample depth of 10 cm. The equation for SOC3 (kg C m-2) was (Resh et al., 2002): SOC3 = (% C3 / 100) x SOCT 27

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

(3.6 mmol h−1 g−1) (Fig. 3). Mean PER activity required removal of 2 outliers and a log transformation to conform to normality and was not dependent upon species (p = 0.15) (Fig. 3). Mean PPO activity required removal of one outlier and a log transformation and was not dependent upon tree species (p = 0.18; Fig. 3). PER and PPO activity were both significantly greater in the more dense 1 x 1 m plots (0.65 and 0.56 mmol h−1 g−1, respectively) than the 2 x 2 m plots (0.35 and 0.21 mmol h−1 g−1, respectively; p = 0.005 and 0.002). Mean NAG activity varied by tree species (p < 0.001), with NAG activity significantly higher in chestnut (1.99 mmol h−1 g−1) and oak (1.91 mmol h−1 g−1) relative to cherry (0.91 mmol h−1 g−1). BG activity did not vary by species (p = 0.24). BG:NAG ratio, indicating the ratio of investment towards enzymes related to C and N acquisition, was significantly affected by tree species (p < 0.0001). BG:NAG ratio was calculated using raw BG and NAG activity data and required removal of two outliers and a log transformation. The back-transformed mean BG:NAG ratio of chestnut (7.9) and oak (10.7) were significantly lower than that of cherry (15.7) (Fig. 3). BG:AP ratio was similar across species (p > 0.05).

Values of % C3-derived SOM where then used in the statistical model with spacing and species as effects. 3. Results 3.1. Leaf litter lignin, tannin, C:N content, and mass loss Initial (T = 0) acid-insoluble compound concentration within litter varied by tree species (p < 0.001), with mean acid-insoluble compound and soluble lignin concentration of oak litter (57.8%) significantly greater than chestnut (56.6%) and cherry (55.2%). At T = 1 y, acid-insoluble compound concentration still varied by species (p < 0.001), with mean acid-insoluble compound and soluble lignin concentration of chestnut litter (43.0%) significantly lower than oak (46.9%), and significantly lower than cherry (52.9%) (Fig. 1). Insoluble compound loss over time was also dependent upon species (p < 0.001), with loss in chestnut litter (13.6%) and oak litter (10.9%) significantly greater than loss by cherry litter (2.3%) (Fig. 1). Initial tannin concentration of litter varied by species (p < 0.001), with mean tannin concentration greatest in cherry (28.5%), followed by oak (16.0%), and lowest in chestnut litter (8.5%). Tannin concentration of litter at T = 1 y required a log transformation to conform to the assumption of normality and was dependent upon species (p < 0.001). The back-transformed mean tannin concentration of at T = 1 y in chestnut litter (6.7%) and oak litter (5.7%) remained significantly lower than cherry litter tannin concentration (10.8%) (Fig. 1). Proportion of tannin loss by mass during the incubation also required a log transformation and was dependent upon species (p < 0.001), where the proportion of tannin loss of chestnut litter (1.8%) was significantly lower than oak litter (10.3%), and significantly lower than cherry litter (17.9%) (Fig. 1). Tannin loss was also significantly greater in litter from the more dense 1 × 1 m plots (11.6%) relative to the 2 × 2 m (8.3%; p < 0.001). Initial litter C:N ratio varied by species (p < 0.001), with mean C:N greatest in chestnut (40.8), followed by cherry (33.0), and lowest in oak litter (26.0). Litter C:N ratio after one year also varied by species (p < 0.001), as chestnut litter C:N (36.0) remained significantly greater than cherry (28.8) or oak litter (27.0). Litter decomposition after one year was dependent upon species (p < 0.001), with mean chestnut litter mass loss (18.2%) significantly greater than both oak (11.6%) and cherry litter mass loss (14.2%) (Fig. 1).

3.4. Linking litter quality, enzyme activity, and litter mass loss N leaching was negatively correlated to litter mass loss (r = −0.40, p = 0.003) and DOC concentration in leachate (r = −0.32, p = 0.02), but was not correlated with NAG activity (r = −0.25, p = 0.07), or change in litter tannin concentration (r = 0.19, p = 0.18) (Fig. 4). Litter mass loss was also negatively correlated with tannin loss in litter during the incubation period (r = −0.28, p = 0.039), but was not correlated to the concentration of acid insoluble compounds or acid soluble lignin (r = −0.17, p = 0.22) or tannin (r = 0.09, p = 0.53) in litter at T = 1 y, or insoluble compounds and acid soluble lignin loss during the incubation period (r = 0.04, p = 0.79). 3.5. Isotopic δ13C and relative SOC3 inputs by tree species Mean δ13C values of litter was significantly different for each tree species (p < 0.001) and were −29.8, −30.1, and −28.9 for cherry, chestnut, and oak, respectively. Mean C4 end-member corn litter had δ13C value of −13.6. In the surface Ap horizon soil, we detected significant differences in δ13C values by tree species, where cherry soil was significantly more depleted in δ13C relative to oak (p = 0.03), though chestnut soil δ13C was not different than either cherry or oak (δ13C = −24.2, −24.1, and −23.2 for cherry, chestnut, and oak, respectively). However, following nearly 10 years of growth at these plantations, across the three species, no significant differences in SOCT were detected in the surface Ap horizon (p > 0.05; mean SOCT = 5679.7, 5824.6, and 5951.2 kg ha−1 for cherry, chestnut, and oak soil, respectively; Fig. 5). Additionally, species did not significantly affect the proportion of SOC3 of the total SOC pool (p > 0.05; mean SOC3 = 3636.2, 3804.3, and 3461.4 kg C ha-1 for cherry, chestnut, and oak soil, respectively). Thus, the proportion of SOC3 in the Ap horizon derived from tree inputs after 10 years were similar at approximately 64.0, 65.3, and 58.2% for cherry, chestnut, and oak soil, respectively (Fig. 5).

3.2. Soil respiration, leachate N and DOC, and POXC Cumulative C loss through respiration during the one-year incubation was not significantly different between species (p = 0.47; Fig. 2). Respiration rates were significantly higher from the less dense 2 × 2 m tree spacing (299.3 g C m−2) relative to the 1 × 1 m plots (259.0 g C m−2; p = 0.005). Cumulative N leaching rates required a logarithmic transformation to adhere to ANOVA’s assumption of normality. N leaching was significantly affected by tree species (p = 0.02). N leaching from chestnut cores (7.8 mg N kg–1 soil) was significantly lower than cherry cores (11.5 mg N kg–1 soil) and oak cores (12.0 mg N kg–1 soil) (Fig. 2). DOC collected from leachate was also significantly different by tree species (p = 0.01). DOC from chestnut (31.4 mg kg-1 soil) was significantly greater than DOC from oak (26.4 mg kg-1 soil), though DOC from cherry cores (29.7 mg kg-1 soil) was not significantly different from either (Fig. 2). No significant differences in soil POXC content by species were detected (p = 0.18; Fig. 2).

4. Discussion Generally, relative to the co-occurring tree species, American chestnut exhibited rapid initial leaf litter decomposition, high organic C availability, and low N availability from the incubated soil and litter experiment. Results from all parameters measured from the incubation experiment do somewhat align with the hypothesized ECM-AM fungal framework, which contends that soils influenced by ECM fungi will have low N availability and high organic C, with N-limitations to microbial activity. However, differences in total C in surface soil were not evident on this site after 10 years growth.

3.3. Nutrient-acquiring extracellular enzyme activity Enzyme activity and significance by tree species varied for each enzyme measured. AP activity was dependent upon species (p = 0.005), with chestnut (4.9 mmol h−1 g−1) and oak (5.5 mmol h−1 g−1) exhibiting significantly greater AP activity than cherry 28

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

Fig. 2. Mean cumulative C loss through respiration (+/−1 SE) over 10 months incubation (top left), mean cumulative N leached (+/−1 SE) (top right), and mean cumulative dissolved organic C (DOC) (+/−1 SE) (bottom left) collected over 10 months and mean soil oxidizable C (POXC) (bottom right) from cores of black cherry, chestnut, and red oak soil and litter material. Bars with different letters are significantly different by Tukey’s HSD test. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

Fig. 3. Mean enzyme activity (mmol h−1 g−1) (+/−1 SE) of AP (top left), PER (top right) PPO (bottom left), and NAG, BG, and the BG:NAG ratio (bottom right) after one year for black cherry, chestnut, and red oak. Bars with different letters are significantly different by Tukey’s HSD test. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

decomposition. In contrast, chestnut leaf litter had similar lignin content and lower tannin content than oak or cherry and more rapid decomposition after one year. The relatively rapid mass loss of chestnut litter is surprising given that chestnut litter also had higher initial and final lignin:N and C:N ratios. Others have shown a strong negative relationship to leaf litter decomposition and these litter quality indices from other tree species (e.g. Melillo et al., 1982; Sun et al., 2018). We

4.1. Leaf litter lignin, tannin, C:N content, and mass loss Chestnut litter had greater initial mass loss and contained less lignin and acid-insoluble compounds after one year than oak or cherry litter and contained less tannins after one year than cherry litter. This does not support the hypothesis that chestnut litter would contain higher concentrations of recalcitrant compounds, resulting in slower 29

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

Fig. 4. Correlations of N mineralized during incubation experiment and litter mass loss (top left), litter C:N ratio (top right), litter lignin:N ratio (mid left), DOC (mid right), and NAG enzyme activity (bottom left). Correlation coefficient and p-value for all species together are reported for each parameter.

suggest that our study timeline of one year and 10–18% mass loss of litter represents only the earliest stages of decomposition, and if monitoring were to continue, it is highly possible that cherry litter decomposition would accelerate to a rate greater than oak or chestnut (Sun et al., 2018). Also, after one year, chestnut had the greatest lignin loss and remaining lignin content was lowest of all species, perhaps suggesting a cultivated soil microbial community capable of degrading lignin compounds from chestnut litter via extracellular enzymes, such as white-rot fungi (Kirk and Farrell, 1987) or other bacteria with hemerelated proteins (Brown et al., 2011). Additionally, given the design of our laboratory incubation experiment that excluded the influence of soil macrofauna and mycorrhizal roots in the decomposition process, it is possible that photodegradation of the litter (instead of microbial decomposition) contributed a significant importance to the mass loss (Austin and Ballaré, 2010). In the photodegradation process, greater lignin content has been shown increase the rate of loss, as lignin is preferentially degraded because it is an effective light-absorbing compound.

Fig. 5. Total soil organic C and soil organic C derived from C3 trees as determined from natural abundance of δ13C isotopes and C concentration and bulk density from surface mineral soil (0–10 cm) from plantation plots. Bars represent means of composite samples from 6 plots per species. Error bars represent +/−1 SE from the mean.

30

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

Chestnut soil and litter produced significantly more DOC in leachate than oak soil and litter, but was similar in DOC production to cherry (Fig. 2). Greater DOC in soil typically represents lower microbial C demand to due limitation of N, as mineralization of C is dependent upon energy demand (Aber, 1992; McDowell et al., 1998). We expected the DOC in leachate from oak and chestnut to be distinct between cherry, supporting results of a study in Moore’s Creek, Indiana, USA, that reported 0.64 mg g−1 in soil influenced by ECM trees compared to 0.32 mg g−1 in AM-dominated soil (Phillips et al., 2013). Our results suggest significant differences in DOC in leachate between the two ECM-associated species, and no differences in DOC in leachate of chestnut relative to cherry, despite its AM association. However, chestnut’s rapid decomposition helps explain the greater DOC produced in leachate, as more C enters the soil as litter decays, and the low inorganic N availability would result in less C mineralization. No statistical differences in POXC in soil from chestnut, oak or cherry were detected. While differences in mean values were relatively large (POXC = 230.7, 284.5, and 253.5 mg kg−1 soil for cherry, chestnut, and oak, respectively), variation in measured values of POXC within species was high in our relatively small number of samples (Fig. 2). These results are similar in scale of POXC differences documented between deciduous species in plantations in the Huitong National Research Station of Forest Ecosystems, in the Hunan providence of China (Wang et al., 2013b). In past studies, POXC has been found to be strongly related to heavier and smaller POC fractions, indicating it reflects a heavily processed, stable pool of C in SOM, and is therefore a suitable parameter for aiding in predicting long-term soil C sequestration and is sensitive to detecting influences of land-use change in C dynamics (Culman et al., 2012). Therefore, while current results are non-significant, it warrants further investigation of the processed C fraction and microbial activity using greater samples sizes and techniques such as the assessment of mineral-associated versus particulate C fractions (MAOM, POM; Haddix et al., 2016).

Litter mass loss was also significantly negatively correlated to initial tannin content and tannin loss from litter. As more tannins were lost from the oak and cherry litter, it is possible the greater flux of tannins entering the soil inhibited further decomposition of litter material (Joanisse et al., 2007). This provides another possible mechanism for the rapid decomposition of chestnut litter relative to oak or cherry. 4.2. Soil respiration, leachate N and DOC, and POXC Importantly, there was no significant species effect on C loss from soils through respiration. While respiration from chestnut cores was not lower than oak or cherry as hypothesized, similarity across species may indicate that the greater DOC and processed C in chestnut soils is not quickly respired and may be available for long-term storage on soil mineral surfaces (Cotrufo et al., 2013). These data support the hypothesis that influence by chestnut would lead to greater long-term SOM accumulation than oak or cherry and aligns with models that predict greater C storage and lesser C turnover below-ground by ECMassociated trees in surface soils (Averill et al., 2014; Taylor et al., 2016). The similarity in respiration is not unusual, as previous studies on C cycling of European species reported oak, sugar maple, and Norway spruce each had mean heterotrophic respiration fluxes of 1.69 μmol m−2 s–1, though beech was significantly lower than both at 1.27 μmol m−2 s–1 (Vesterdal et al., 2012). Additionally, our measures were performed in a temperature- and moisture-controlled setting, removing potential canopy, climate, and macroinvertebrate influences that might occur in the field, and resulting in more uniform respiration activity across tree species. Chestnut cores lost significantly less mineral N than oak or cherry, which supports the hypothesis that soil influenced by chestnut would have lower N mineralization rates and aligns with the mycorrhizalmediated C and N cycling model that predicts greater surface soil C and lower inorganic N beneath ECM trees compared to AM (Averill et al., 2014). The difference between chestnut and cherry found here follows a similar pattern to a study in Harvard Forest, MA, USA that found much higher N mineralization rates of both bulk and rhizosphere soil influenced by sugar maple (Acer saccharum) than the ECM species American beech (Fagus grandifolia) (Brzostek et al., 2013). However, because tannin concentrations in chestnut litter were not higher than that of cherry or oak, the mechanism driving this is not clear, but is likely due to the low initial N content of the litter. It is also possible that the HCl – butanol method used here to quantify tannins, while commonly used for measuring PAs or condensed tannins, may not be the most indicative measure of species-specific tannins that differ in molecular structure from the common standard. This may result in over- or underestimation of the amount of tannins depending on the species. Thus, it is not really possible to compare the absolute tannin values between species based on the HCl - butanol method when using the same standard rather than tree species-specific standards for each species litter. Species-specific differences in molecular structure of PAs may also differentially influence microbial activity (Hättenschwiler and Vitousek, 2000). For example, Schimel et al. (1998) reported that decomposition and N mineralization was inhibited by additions of PA from balsam alder, but not by PA from thinleaf alder. N leaching rates were negatively correlated with litter C:N and lignin:N ratios, NAG enzyme activity, DOC leachate, and litter decomposition (Fig. 4). The relationship between DOC and N mineralization in soil is well-established and is thought to be due to microbial energy constraints in N-limited systems (Aber, 1992; McDowell et al., 1998). This indicates that the lack of inorganic N in chestnut-dominated soils reduces the demand for C uptake by microbes, thus reducing C mineralization. The relationship here between litter lignin:N and N leaching here also supports past reports (e.g. Scott and Binkley, 1997), where high lignin:N ratios are correlated to low N mineralization. Chestnut had the highest lignin:N values and lowest N leaching relative to oak and cherry.

4.3. Nutrient-acquiring extracellular enzyme activity Extracellular enzyme activities indicate that chestnut soil has greater microbial N- and P-limitations. In chestnut soil, NAG and AP activity were greater than cherry, and non-significantly lower PER and PPO than oak or cherry. This, coupled with low tannin content in chestnut litter, does not support the hypothesis that chestnut soils have lower overall enzyme activity than both oak and cherry resulting from increased concentrations of tannins inhibiting enzyme activity (Adamczyk et al., 2017). However, that chestnut and oak have greater AP activity may be due to ECM fungi’s ability to release AP enzymes, whereas in AM species, AP is released by roots and free-living microbes (Phillips and Fahey, 2006). The BG:NAG ratio of chestnut and oak soil was significantly lower than cherry (Fig. 3), indicating the microbial community was more Nlimited in chestnut and oak soil, and microbes were investing in producing more NAG enzymes to acquire N. Typically, under N-limited conditions, C use efficiency (CUE) of microbial communities will decrease, and microbes will respire more C per unit N acquired in an effort to get what little N is available (Manzoni et al., 2012). However, an increase in respiration in chestnut soils was not evident relative to oak or cherry, indicating that microbial communities associated with chestnut have a relatively high CUE and extracellular enzymes in chestnut soils break down more N-containing compounds (NAG) relative to simple sugars (BG), and is C-efficient in acquiring N (i.e. indicative of high N use efficiency (NUE)). This is supported by relatively low N mineralization, high DOC, rapid initial leaf litter decomposition, and moderate respiration from chestnut, as the microbial community associated with chestnut are efficient to acquire N while not increasing C respiration loss, resulting in greater DOC from chestnut soil (rapid decomposition, low N mineralization = high NUE) (Mooshammer et al., 2014). 31

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

4.4. Isotopic δ13C and relative SOC3 inputs by tree species

loss through respiration, relative to oak or cherry. Overall, the soil ecosystem from chestnut plots indicates an accumulating microbial biomass despite apparent N-limitation, reflective of greater SOM accumulation potential relative to black cherry and red oak. The 13C isotopic analysis of new SOC3 accumulation in field plots indicate no differences as a function of tree species, likely as a function of the relatively young age of the plantation. We conclude that greater fluxes of inorganic N may have occurred from forested ecosystems as chestnut declined, and suggest potential for enhanced C storage belowground over the long-term if chestnut is successfully reintroduced throughout its native range.

The proportion of SOC3 fraction of C derived from tree species in these field plots do not support our hypothesis that soil influenced by chestnut would have greater accumulation of SOC in surface soil relative to other tree species, especially AM-associated black cherry. It is not surprising that the SOCT pool was not different between species, given the size of the C pool and the amount of time it takes for SOC to form, though we did expect to see differences in the SOC3 pool. Similarities of C3-derived C may be attributed to the young age of the trees at this plantation (< 10 years). In a study of SOC storage beneath different ages of Miscanthus plantations, it was reported that after 9years of growth, SOC in Miscanthus plots were not different than reference plots with C3 plants (91–92 t C ha−1). After 16 years of growth, however, SOC in Miscanthus was significantly greater at 106 t C ha−1 (Hansen et al., 2004). We might expect forest trees to take even longer to influence soils differentially given the differences in life span and rooting and litter deposition relative to grasses. Additionally, in the incubation study, we document significant differences in inorganic N and DOC beneath tree species in fine-textured, well-buffered silt loam soils, which are relatively resistant to biogeochemical change. We might expect even larger influences of tree species on C and N dynamics in areas of coarser-textured soils, similar to the findings by Rhoades (2007). As such, given the relative homogeneity of SOCT and SOC3 values we report here using the δ13C mixing-model approach, we must place a caveat on the potential of chestnut to enhance below-ground C storage suggested by the incubation study (low respiration, high DOC, and rapid decomposition/potentially greater microbially processed C) to say it may take many years to realize species-influenced differences in the larger context of total soil organic C.

Author contributions CK conceived and designed the research; GS performed the experiments; CK and GS analyzed the data; GS and CK wrote and edited the manuscript. Declaration of interest None Acknowledgements Funding: This work was supported by the USDA NIFA McIntireStennis [award number 1008506]. The authors thank Douglass Jacobs and Brian Beheler of Purdue University for access to tree plantations and field assistance. We thank Timothy Driscoll, Zachary Freedman, and Ben Andoh for access to analytical labs and William Peterjohn and Edward Brzostek for comments and edits on an earlier version of this manuscript.

4.5. Potential limitations of the study and recommendations for future monitoring

References

This study utilizes the hypothesized framework of AM-ECM driven nutrient cycling, but does not explicitly test differences between AMand ECM-associated tree species, as replication within AM or ECM trees was not present (AM) or very small (ECM). While making general statements about the influence of the two mycorrhizal types is impossible, our data adds to the growing body of information of treespecies influences on nutrient cycling, particularly related to American chestnut. Additionally, the study design of the lab-incubated soil and litter material excludes important ecosystem components that may strongly influence decomposition and mineralization, including the exclusion of fine roots, soil fauna, and differences in micro-climate beneath different tree species. Especially regarding fine root decay, Sun et al. (2018) recently showed that fine root decay could be significantly more important than leaf litter decay in determining soil functions and C storage, and that fine roots can have highly differing turnover depending on tree species and the corresponding characteristics of the roots, particularly the non-lignin C chemistry of the roots. Results of the Sun et al. (2018) study indicate that ECM-associated fine roots decayed faster than AM roots after 2 years. Future monitoring at these plots and other locations should include measures that encompass the importance of fine root characteristics and decomposition rates, as well as other soil fauna and micro-climate influences on C turnover.

Aber, J.D., 1992. Nitrogen cycling and nitrogen saturation in temperate forest ecosystems. Trends Ecol. Evol. (Amst.) 7 (7), 220–224. Adamczyk, B., Simon, J., Kitunen, V., Adamczyk, S., Smolander, A., 2017. Tannins and their complex interaction with different organic nitrogen compounds and enzymes: old paradigms versus recent advances. ChemistryOpen. 6 (5), 610–614. Austin, A.T., Ballaré, C.L., 2010. Dual role of lignin in plant litter decomposition in terrestrial ecosystems. Proc. Natl. Acad. Sci. U.S.A. 107 (10), 4618–4622. Averill, C., Turner, B.L., Finzi, A.C., 2014. Mycorrhiza-mediated competition between plants and decomposers drives soil carbon storage. Nature 505 (7484), 543. Braun, E.L., 1950. Deciduous Forests of Eastern North America. Deciduous forests of Eastern North America. Blakiston Co. Philadelphia, Pennsylvania. Brown, M.E., Walker, M.C., Nakashige, T.G., Iavarone, A.T., Chang, M.C., 2011. Discovery and characterization of heme enzymes from unsequenced bacteria: application to microbial lignin degradation. J. Am. Soc. Brew. Chem. 133 (45), 18006–18009. Brzostek, E.R., Greco, A., Drake, J.E., Finzi, A.C., 2013. Root carbon inputs to the rhizosphere stimulate extracellular enzyme activity and increase nitrogen availability in temperate forest soils. Biogeochem 115 (1-3), 65–76. Chen, H., Sang, S., 2014. Biotransformation of tea polyphenols by gut microbiota. J. Funct. Foods 7, 26–42. Christopher, S.F., Page, B.D., Campbell, J.L., Mitchell, M.J., 2006. Contrasting stream water NO3− and Ca2+ in two nearly adjacent catchments: the role of soil Ca and forest vegetation. Glob. Change Biol. Bioenergy 12 (2), 364–381. Cotrufo, M.F., Wallenstein, M.D., Boot, C.M., Denef, K., Paul, E., 2013. The Microbial Efficiency‐Matrix Stabilization (MEMS) framework integrates plant litter decomposition with soil organic matter stabilization: do labile plant inputs form stable soil organic matter? Glob. Change Biol. Bioenergy 19 (4), 988–995. Craig, M.E., Turner, B.L., Liang, C., Clay, K., Johnson, D.J., Phillips, R.P., 2018. Tree mycorrhizal type predicts within‐site variability in the storage and distribution of soil organic matter. Glob. Change Biol. Bioenergy 24 (8), 3317–3330. Culman, S.W., Snapp, S.S., Freeman, M.A., Schipanski, M.E., Beniston, J., Lal, R.…, Lee, J., 2012. Permanganate oxidizable carbon reflects a processed soil fraction that is sensitive to management. Soil Sci. Soc. Am. J. 76 (2), 494–504. DeForest, J.L., 2013. Optimized Lab Protocols for Testing Soils. Department of Environmental and Plant Biology. Ohio University, Athens, OH. Ellison, A.M., Bank, M.S., Clinton, B.D., Colburn, E.A., Elliott, K., Ford, C.R., Foster, D.R., Kloeppel, B.D., Knoepp, J.D., et al., 2005. Loss of foundation species: consequences for the structure and dynamics of forested ecosystems. Front. Ecol. Environ. 3 (9), 479–486. Finzi, A.C., Van Breemen, N., Canham, C.D., 1998. Canopy tree–soil interactions within temperate forests: species effects on soil carbon and nitrogen. Ecol. Appl. 8 (2), 440–446.

5. Conclusion The results of our short-term incubation study indicate that soils influenced by American chestnut had lower inorganic N availability compared to oak and cherry, and chestnut litter underwent rapid litter decomposition and produced high DOC in leachate. In addition, the microbial community within soils influenced by chestnut litter is seemingly efficient in its use of C to acquire N, resulting in no increased C 32

Pedobiologia - Journal of Soil Ecology 75 (2019) 24–33

G.W. Schwaner and C.N. Kelly

Moreira-Vilar, F.C., de Cássia Siqueira-Soares, R., Finger-Teixeira, A., de Oliveira, D.M., Ferro, A.P., da Rocha, G.J., et al., 2014. The acetyl bromide method is faster, simpler and presents best recovery of lignin in different herbaceous tissues than Klason and thioglycolic acid methods. PLoS One 9 (10), e110000. Nelson, T.C., 1955. Chestnut replacement in the southern highlands. Ecology 36, 352–353. Peterjohn, W.T., Foster, C.J., Christ, M.J., Adams, M.B., 1999. Patterns of nitrogen availability within a forested watershed exhibiting symptoms of nitrogen saturation. For. Ecol. Manage. 119 (1-3), 247–257. Phillips, R.P., Fahey, T.J., 2006. Tree species and mycorrhizal associations influence the magnitude of rhizosphere effects. Ecology 87 (5), 1302–1313. Phillips, R.P., Brzostek, E., Midgley, M.G., 2013. The mycorrhizal‐associated nutrient economy: a new framework for predicting carbon–nutrient couplings in temperate forests. New Phytol. 199 (1), 41–51. Preston, C.M., 1999. Condensed tannins of salal (Gaultheria shallon Pursh): a contributing factor to seedling “growth check” on northern Vancouver Island? Plant Polyphenols 2. Springer, Boston, MA, pp. 825–841. Preston, C.M., Trofymow, J.A., Sayer, B.G., Niu, J., 1997. 13C nuclear magnetic resonance spectroscopy with cross-polarization and magic-angle spinning investigation of the proximate analysis fractions used to assess litter quality in decomposition studies. Can. J. Bot. 75, 1601–1613. Resh, S.C., Binkley, D., Parrotta, J.A., 2002. Greater soil carbon sequestration under nitrogen-fixing trees compared with Eucalyptus species. Ecosystems 5 (3), 217–231. Rhoades, C.C., 2007. The influence of American chestnut (Castanea dentata) on nitrogen availability, organic matter and chemistry of silty and sandy loam soils. Pedobiologia 50 (6), 553–562. Rhoades, C., Loftis, D., Lewis, J., Clark, S., Service, U.F., 2009. The influence of silvicultural treatments and site conditions on American chestnut (Castanea dentata) seedling establishment in eastern Kentucky, USA. Forest Ecology and Management 258 (7), 1211–1218. Rosenberg, C., 2010. Foliar Chemical Analysis, Decomposition, and Effects on Nutrient Cycling of American Chestnut and Its Hybrids. MS Thesis. Ohio University. Schimel, J.P., Cates, R.G., Ruess, R., 1998. The role of balsam poplar secondary chemicals in controlling soil nutrient dynamics through succession in the Alaskan taiga. Biogeochem 42 (1-2), 221–234. Schulp, C.J., Nabuurs, G.J., Verburg, P.H., de Waal, R.W., 2008. Effect of tree species on carbon stocks in forest floor and mineral soil and implications for soil carbon inventories. For. Ecol. Manage. 256 (3), 482–490. Scott, N.A., Binkley, D., 1997. Foliage litter quality and annual net N mineralization: comparison across North American forest sites. Oecologia 111 (2), 151–159. Sinsabaugh, R.L., 2008. Stoichiometry of soil enzyme activity at global scale. Ecol. Lett. 11, 1252–1264. Sinsabaugh, R.L., Carreiro, M.M., Repert, D.A., 2002. Allocation of extracellular enzymatic activity in relation to litter composition, N deposition, and mass loss. Biogeochemistry 60 (1), 1–24. Soil Survey Staff, 2016. Natural Resources Conservation Service, United States Department of Agriculture. Web Soil Survey. Available online at the following link: https://websoilsurvey.sc.egov.usda.gov/ Accessed 06/04/. Sun, T., Hobbie, S.E., Berg, B., Zhang, H., Wang, Q., Wang, Z., Hättenschwiler, S., 2018. Contrasting dynamics and trait controls in first-order root compared with leaf litter decomposition. Proc Nat Acad Sci USA. 115 (41), 10392–10397. Taylor, M.K., Lankau, R.A., Wurzburger, N., 2016. Mycorrhizal associations of trees have different indirect effects on organic matter decomposition. J. Ecol. 104 (6), 1576–1584. Vesterdal, L., Elberling, B., Christiansen, J.R., Callesen, I., Schmidt, I.K., 2012. Soil respiration and rates of soil carbon turnover differ among six common European tree species. For. Ecol. Manage. 264, 185–196. Vesterdal, L., Clarke, N., Sigurdsson, B.D., Gundersen, P., 2013. Do tree species influence soil carbon stocks in temperate and boreal forests? For. Ecol. Manage. 309, 4–18. Volkoff, B., Cerri, C.C., 1987. Carbon isotopic fractionation in subtropical Brazilian grassland soils. Comparison with tropical forest soils. Plant Soil 102 (1), 27–31. Wang, G.G., Knapp, B.O., Clark, S.L., Mudder, B.T., 2013a. The Silvics of Castanea dentata (Marsh.) Borkh. American Chestnut, Fagaceae (Beech Family). US Forest Service, Southern Research Station. General Technical Report SRS-173. Wang, Q., Xiao, F., Zhang, F., Wang, S., 2013b. Labile soil organic carbon and microbial activity in three subtropical plantations. Forestry 86 (5), 569–574. Yin, H., Wheeler, E., Phillips, R.P., 2014. Root-induced changes in nutrient cycling in forests depend on exudation rates. Soil Biol. Biochem. 78, 213–221. Youngs, R.L., 2000. "A Right Smart Little Jolt": Loss of the Chestnut and a Way of Life. J For. 98 (2), 17–21. Zhang, B., Oakes, A.D., Newhouse, A.E., Baier, K.M., Maynard, C.A., Powell, W.A., 2013. A threshold level of oxalate oxidase transgene expression reduces Cryphonectria parasitica-induced necrosis in a transgenic American chestnut (Castanea dentata) leaf bioassay. Transgenic Res. 22 (5), 973–982.

Gauthier, M.M., Zellers, K.E., Löf, M., Jacobs, D.F., 2013. Inter-and intra-specific competitiveness of plantation-grown American chestnut (Castanea dentata). For. Ecol. Manage. 291, 289–299. Giardina, C.P., Ryan, M.G., Hubbard, R.M., Binkley, D., 2001. Tree species and soil textural controls on carbon and nitrogen mineralization rates. Soil Sci Soc Am J . 65 (4), 1272–1279. Haddix, M.L., Paul, E.A., Cotrufo, M.F., 2016. Dual, differential isotope labeling shows the preferential movement of labile plant constituents into mineral‐bonded soil organic matter. Glob. Change Biol. Bioenergy 22 (6), 2301–2312. Hagerman, A.E., Butler, L.G., 1981. The specificity of proanthocyanidin-protein interactions. J. Biol. Chem. 256 (9), 4494–4497. Hansen, E.M., Christensen, B.T., Jensen, L.S., Kristensen, K., 2004. Carbon sequestration in soil beneath long-term Miscanthus plantations as determined by 13C abundance. Biomass Bioenergy 26 (2), 97–105. Hättenschwiler, S., Vitousek, P.M., 2000. The role of polyphenols in terrestrial ecosystem nutrient cycling. Trends Ecol. Evol. (Amst.) 15 (6), 238–243. Hebard, F.V., 2006. The backcross breeding program of the American Chestnut Foundation. J. Am. Chestnut Foundation 19, 55–77. Ibáñez, A.B., Bauer, S., 2014. Downscaled method using glass microfiber filters for the determination of Klason lignin and structural carbohydrates. Biomass Bioenergy 68, 75–81. Jacobs, D.F., 2007. Toward development of silvical strategies for forest restoration of American chestnut (Castanea dentata) using blight-resistant hybrids. Biol. Conserv. 137 (4), 497–506. Jacobs, D.F., Severeid, L.R., 2004. Dominance of interplanted American chestnut (Castanea dentata) in southwestern Wisconsin, USA. For. Ecol. Manage. 191 (1-3), 111–120. Jacobs, D.F., Dalgleish, H.J., Nelson, C.D., 2013. A conceptual framework for restoration of threatened plants: the effective model of American chestnut (Castanea dentata) reintroduction. New Phytol. 197 (2), 378–393. Jandl, R., Lindner, M., Vesterdal, L., Bauwens, B., Baritz, R., Hagedorn, F., et al., 2007. How strongly can forest management influence soil carbo sequestration? Geoderma 137 (3-4), 253–268. Joanisse, G.D., Bradley, R.L., Preston, C.M., Munson, A.D., 2007. Soil enzyme inhibition by condensed litter tannins may drive ecosystem structure and processes: the case of Kalmia angustifolia. New Phytol. 175 (3), 535–546. Kirk, T.K., Farrell, R.L., 1987. Enzymatic "combustion": the microbial degradation of lignin. Ann Rev Microbiol. 41 (1), 465–501. Knelman, J.E., Graham, E.B., Ferrenberg, S., Lecoeuvre, A., Labrado, A., Darcy, J.L., et al., 2017. Rapid shifts in soil nutrients and decomposition enzyme activity in early succession following forest fire. Forests 8 (9), 347. Lin, G., McCormack, M.L., Ma, C., Guo, D., 2017. Similar below-ground carbon cycling dynamics but contrasting modes of nitrogen cycling between arbuscular mycorrhizal and ectomycorrhizal forests. New Phytol. 213, 1440–1451. Lorenz, K., Preston, C.M., Raspe, S., Morrison, I.K., Feger, K.H., 2000. Litter decomposition and humus characteristics in Canadian and German spruce ecosystems: information from tannin analysis and 13C CPMAS NMR. Soil Biol. Biochem. 32 (6), 779–792. Lovett, G.M., Weathers, K.C., Arthur, M.A., 2002. Control of nitrogen loss from Forested watersheds by soil carbon: nitrogen ratio and tree species composition. Ecosystems 5 (7), 0712–0718. Lovett, G.M., Weathers, K.C., Arthur, M.A., Schultz, J.C., 2004. Nitrogen cycling in a northern hardwood forest: do species matter? Biogeochem 67 (3), 289–308. Magill, A.H., Aber, J.D., Berntson, G.M., McDowell, W.H., Nadelhoffer, K.J., Melillo, J.M., Steudler, P., 2000. Long-term nitrogen additions and nitrogen saturation in two temperate forests. Ecosystems 3 (3), 238–253. Manzoni, S., Taylor, P., Richter, A., Porporato, A., Ågren, G.I., 2012. Environmental and stoichiometric controls on microbial carbon‐use efficiency in soils. New Phytol. 196 (1), 79–91. McCament, C.L., McCarthy, B.C., 2005. Two-year response of American chestnut (Castanea dentata) seedlings to shelterwood harvesting and fire in a mixed-oak forest ecosystem. Can. J. Res. 35 (3), 740–749. McClaugherty, C.A., Pastor, J., Aber, J.D., Melillo, J.M., 1985. Forest litter decomposition in relation to soil nitrogen dynamics and litter quality. Ecology 66 (1), 266–275. McDowell, W.H., Currie, W.S., Aber, J.D., Yang, Y., 1998. Effects of chronic nitrogen amendments on production of dissolved organic carbon and nitrogen in forest soils. Biogeochemical Investigations at Watershed, Landscape, and Regional Scales. Springer, Dordrecht, pp. 175–182. McEwan, R.W., Keiffer, C.H., McCarthy, B.C., 2006. Dendroecology of American chestnut in a disjunct stand of oak chestnut forest. Can. J. Res. 36 (1), 1–11. Melillo, J.M., Aber, J.D., Muratore, J.F., 1982. Nitrogen and lignin control of hardwood leaf litter decomposition dynamics. Ecology 63 (3), 621–626. Mooshammer, M., Wanek, W., Hämmerle, I., Fuchslueger, L., Hofhansl, F., Knoltsch, A., et al., 2014. Adjustment of microbial nitrogen use efficiency to carbon: nitrogen imbalances regulates soil nitrogen cycling. Nature Comm. 5, 3694.

33