- Email: [email protected]
Does species richness of subtropical tree leaf litter affect decomposition, nutrient release, transfer and subsequent uptake by plants?
Soil Biology & Biochemistry 115 (2017) 44e53
Contents lists available at ScienceDirect
Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio
Does species richness of subtropical tree leaf litter affect decomposition, nutrient release, transfer and subsequent uptake by plants? Katrin N. Leppert a, *, Pascal A. Niklaus b, Michael Scherer-Lorenzen a a b
University of Freiburg, Faculty of Biology, Geobotany, Schaenzlestr. 1, 79104 Freiburg, Germany University of Zurich, Institute of Evolutionary Biology and Environmental Studies, Winterthurerstrasse 190, 8057 Zurich, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 27 March 2017 Received in revised form 18 July 2017 Accepted 3 August 2017
During leaf litter decomposition, nutrients are released, can be transferred among different litter species, are metabolized by soil organisms and are taken up by plants again. However, it remains unclear to which extent leaf litter species richness affects these processes of nutrient cycling, and whether effects on one of those processes propagate to the subsequent one. We established a common garden decomposition experiment in a Chinese subtropical secondary forest, to trace two essential nutrients during decomposition and their uptake by plants along a litter species diversity gradient. Unlabelled, and 15N and Li (as surrogate for K) labelled leaves of three native tree species were used to create replicated 1-, 2and 3-species mixtures, each with one species labelled per mixture. Litter mixtures were placed in mesocosms with one growing herbaceous phytometer plant. Over six months, litter and phytometer plants of each mixture were sampled at four points in time and the different process steps of nutrient dynamics were determined. Our results showed species and nutrient specific decomposition dynamics, which propagated through the processes of mass loss, nutrient release and transfer among species, and nutrient uptake dynamics of phytometer plants. However, we found no litter species diversity effects along the different litter decomposition processes. Rather specific diversity effects occurred in few cases at different points in time for mass loss, Li release and transfer dynamics. These effects were not caused by nutrient transfer from labelled to unlabelled litter, suggesting that species identity effects on decomposer dynamics may outweigh effects of nutrient transfer among litter species in mixtures. Further, the observed litter species diversity effects did not affect the 15N uptake of phytometer plants. Hence, the influence of species diversity on nutrient cycling and plant available nutrient stocks is mainly determined by the amount and variety of chemical compounds that different species exhibit and release to the soil. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Litter species diversity Leaf litter decomposition Nutrient dynamics Tracer elements BEF-China Subtropical forests
1. Introduction The significance of plant biodiversity for terrestrial ecosystem functioning and the delivery of ecosystem services has been pointed out over the last decades, emphasizing a strong biotic control over ecological processes (e.g. Hooper et al., 2005; Cardinale et al., 2012; Tilman et al., 2014). Initially focusing on the consequences of biodiversity loss on productivity, many studies now show that plant species diversity and functional composition of communities strongly affect a large variety of ecosystem properties
* Corresponding author. E-mail address: [email protected] (K.N. Leppert). http://dx.doi.org/10.1016/j.soilbio.2017.08.007 0038-0717/© 2017 Elsevier Ltd. All rights reserved.
and processes. Among these, decomposition of dead organic matter is a key function as it affects nutrient cycling, soil formation and carbon storage (Johnson and Todd, 1998; Berg and McClaugherty, 2008). A range of experiments has evaluated the effects of plant diversity on decomposition of leaf litter, demonstrating highly species-specific litter decomposition rates that depend on litter quality traits such as C:N or lignin:N ratios, secondary metabolites or morphological characteristics (e.g. Berg et al., 1992; Cardisch and rez-Harguindeguy et al., 2000). Decomposition rates Giller, 1996; Pe of litter species in mixtures often differ from the average ones in their component monocultures, i.e. they are non-additive (e.g. Seastedt, 1984; Chapman et al., 1988; Ball et al., 2008). Thus, some leaves decompose faster (synergistic effects) or slower
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
(antagonistic effects) than one would expect from respective monoculture decomposition rates or show idiosyncratic responses (non-linear richness effects) (Gartner and Cardon, 2004; €ttenschwiler et al., 2005a). Litter species richness effects can Ha be caused by (1) an increased attractiveness of low quality litter as a food source for decomposers if mixed with high quality litter, via nutrient transfer among leaves of different species, (2) variations of species-specific litter compounds with either stimulating or inhibiting properties, (3) alteration of microclimate and habitat structures through differences in leaf shapes and toughness, and (4) changes in the abundance or activity of certain decomposer guilds, and (5) feedback mechanisms among different decomposer guilds €ttenschwiler et al., 2005a). (reviewed by Ha Seeing litter decomposition simply as the rate of mass loss or CO2-production, as done in most studies, does not adequately reflect the underlying complex, time-dependent phenomena such as leaching of soluble compounds, physical breakup, and biotic shredding, enzymatic breakage of complex compounds, release and transfer of elements among different litter types through leaching and decomposers (e.g. reviewed in Cardisch and Giller, 1996; Berg and McClaugherty, 2008). Potential diversity effects on some of these individual processes, such as the nutrient transfer among litter species in mixtures or the subsequent nutrient uptake by plants, have been rarely considered and remain unclear. Nutrient concentrations of decomposing litter species in mixtures can increase or decrease compared to their respective monocultures (e.g. Briones and Ineson, 1996; McTiernan et al., 1997; Kaneko and Salamanca, 1999). One of the main potential mechanisms is the transport of nutrients among different litter species by leaching and diffusion (Fyles and Fyles, 1993; McArthur, 1994; Briones and Ineson, 1996) or transport by fungal hyphae (Frey et al., 2003; Tiunov, 2009; Lummer et al., 2012) which can improve decomposability of more recalcitrant litter species. Further, nutrients may partially originate from external sources (e.g. Peterson and Rolfe, 1982), such as fungal colonization (Caner et al., 2004) and fungal transport of deposited N from soil (Fahey et al., 2011). Whether subsequent richness effects on litter mass loss and nutrient release rates affect the nutrient uptake by plants still remains unclear. The study of such detailed decomposition processes and their underlying mechanisms is methodological challenging and ideally requires 1) a separation of the leaf litter according to species and 2) the use of isotope tracer elements to determine the nutrient sources and sinks. Common tracer elements to follow nutrients through the plant-soil system are stable isotopes like 15N, or rare elements which serve no apparent vital biological function but have similar chemical properties as other elements essential for plant growth. Lithium (Li), for example, can be used to quantify plant uptake of potassium (K) (Gockele et al., 2014) as they share Kþ transport carriers in roots (Fitter, 1986). By using tracers it is also possible to detect and quantify element transfer from labelled (donor) to €ttenschwiler, unlabelled (acceptor) leaf litter (e.g. Schimel and Ha 2007; Lummer et al., 2012). For example, Lummer et al. (2012) revealed nitrogen transfer through fungal hyphae from nitrogenpoor to nitrogen-rich leaf litter. These results challenge the traditional view of nitrogen transfer being one-directional from nitrogen-rich to nitrogen-poor litter species, and suggest that nutrient transfer mechanisms among litter species are not completely understood. Tracer elements have also been used to follow the transport of nutrients from decomposing leaves to soil and plants (e.g. Zeller et al., 2000, 2001) including the uptake by decomposer organisms (e.g. Caner et al., 2004). Zeller et al. (2000, 2001) documented a 1:1 relationship between litter mass loss and nitrogen release. After two years of litter decomposition, 35% of the nitrogen remained in the recalcitrant part of the litter, 50% reached the topsoil and about 2e4% was incorporated into 15e50
45
year old beech trees (Zeller et al., 2001). However, these studies focused on pure or low-diverse litter mixtures only, neglecting litter mixtures, and macro-nutrients other than nitrogen. Although these studies provide important insights into mechanisms of nutrient dynamics, they are thus only of limited use for the understanding of litter diversity effects on decomposition and nutrient cycling, especially within litter mixtures. Here, we specifically addressed the question of how litter species richness affects overall litter decomposition rates (i.e. mass loss), nutrient release and transfer among leaves in mixtures and the subsequent uptake by plants. We conducted a field litter decomposition experiment in the subtropics of south-east China and focused on the macro-nutrients N and K, which were represented by the tracers 15N and Li, to include one immobile and one highly mobile element in leaf litter (Briones and Ineson, 1996; Yang et al., 2004). 15N and Li labelled and unlabelled, senescent leaves of three native tree species were incubated in situ as monocultures, 2and 3-species mixtures in mesocosms and decomposition rates and tracer contents in the labelled and unlabelled litter species were followed over 24 weeks. We further used a phytometer approach to quantify nutrient transfer dynamics from leaves into plants. We tested three hypotheses: (i) Temporal dynamics for mass loss, nutrient release, transfer and uptake by plants are highly species- and nutrient specific. Due to strong functional links among these processes involved in decomposition, species identity effects will be consistent for these processes. (ii) Litter species richness positively affects litter mass loss and nutrient dynamics. These richness effects will propagate to subsequent processes, and will therefore also be found for nutrient release and uptake by phytometer plants. (iii) In case of litter species richness effects on litter mass loss, nutrient release and the uptake by indicator plants we expect richness effects also on nutrient transfer among litter species. This expectation is based on the fact that nutrient transfer among litter species has been hypothesized to be a key mechanism explaining diversity effects on decomposition processes €ttenschwiler et al., 2005a). (Ha 2. Materials and methods 2.1. Study site We conducted a litter decomposition experiment from Apr. (2012) to Oct. 2012 in Xingangshan, Jiangxi province in south-east China (29.1200 N, 117.91” E). The climate is subtropical with warm, wet summers and cold, dry winters. Mean annual temperature is 16.7 C and annual precipitation averages 1821 mm (Yang et al., 2013). The field site was situated on a mountain ridge in a small, natural mixed forest stand with Castanopsis calesii (HEMSLEY) HAYATA, Loropetalum chinense (R. BROWN) OLIVER, Eurya muricata DUNN, Lithocarpus glaber (THUNBERG) NAKAI and Castanopsis eyrei (CHAMPION EX BENTHAM) TUTCHER as dominant tree species (Eichenberg, personal communication). Besides few plants of Carex spec., no herb layer existed. The site is situated on an Endoleptic Cambisol (IUSS Working Group WRB, 2015) on silt loam (24% sand, 50% silt and 26% clay) with a mean pH in the topsoil layer (0e5 cm) of 4.7 (Seitz et al., 2015). 2.2. Production of labelled leaf litter We established a nursery with the native tree species Sapindus saponaria LINNAEUS, Quercus acutissima CARRUTHERS and Schima superba GARDNER & CHAMPION to produce both labelled and unlabelled leaf litter. A total of approx. 6000 3e6 year old trees of all three species were planted in spring 2011. One third of the trees was watered twice with 200 ml 0.001 M 15NH4Cl and 0.236 M LiCl,
46
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
respectively. Labelled and unlabelled trees were separated by natural borders or coarse meshed fences to prevent mixing of labelled and unlabelled leaf litter. Freshly fallen leaves were collected weekly from the ground in autumn 2011, omitting damaged ones. Leaves were pooled by species and label status (labelled and unlabelled), and dried at 40 C. Mean initial tracer concentrations in labelled leaf litter indicated a successful tracer uptake for almost all species (Table 1). Only mixtures with labelled Q. acutissima leaves were excluded from Li transfer statistics due to very low initial Li concentrations (Tables 2 and 3 comment ‘1’). 2.3. Experimental design and data collection We created a total of seven different litter species combinations, including three monocultures, three 2- and one 3-species mixture. Each litter mixture was realized with one species labelled per mixture, resulting in twelve mixtures with different label combinations which were replicated in three blocks (36 plots in total). Each mixture comprised 1/3 labelled and 2/3 unlabelled leaves to achieve similar total 15N and Li contents for each labelled species per mixture. For example, monocultures of S. superba contained 1/3 labelled and 2/3 unlabelled litter of this species; a 2-species mixture of labelled S. superba and unlabelled S. saponaria leaves contained 1/3 labelled and 1/6 unlabelled S. superba litter and 1/2 unlabelled S. saponaria litter; a 3-species mixture with labelled S. superba litter contained 1/3 labelled S. superba, 1/3 unlabelled S. saponaria and 1/3 unlabelled Q. acutissima litter (Fig. 1). These litter mixtures were placed in mesocosms made of PVC (15 cm diameter x 12.5 cm height, lowered 2 cm into the soil), allowing direct leaf-soil contact. Every mesocosm was filled with 5.1 g of the respective litter mixture. One phytometer plant (Carex sp.), originating from the same field site, was transplanted in the centre. The plants were excavated, dry leaves removed, and roots carefully cleaned with water prior to planting. Mesocosms were covered with a 1.5 cm mesh to prevent litter input from surrounding trees. Each plot contained four mesocosms (temporal replicates) of the same species mixture and covered an area of 45 45 cm. Overall, there were a total of 144 mesocosms. The position of all plots within the three blocks were chosen by randomization to account for a possible heterogeneity in soil conditions due to differences of the overstory species composition.
powder (Mixer Mill MM 200, Retsch GmbH, Haan, Germany) and analysed for 15N and Li contents. Total tracer pools were calculated by multiplying tracer concentrations with the total litter weight of the species, including the part estimated in the unsorted fraction. Each phytometer plant was harvested, freed from dead plant material at the base and analysed for its 15N content. Due to small sample volumes, Li could not be determined in the phytometers. First order litter decomposition rate constants k (yr1) for plot and species specific litter mass loss were calculated by fitting Xt ¼ X0 ekt to the data, where X0 and Xt are litter mass at time 0 and t, respectively. The fit of the function was inspected visually and generally matched the data well. 15 N was analysed with an isotope-ratio mass spectrometer (Thermo Fisher Scientific, Finnigan DELTAplus, Waltham, USA) at the Centre for Biological Systems Analysis in Freiburg. Total tracer 15 N contents (mmol sample1) are reported as 15Nexcess and were calculated with the formula 15 Nexcess ¼ (atom%15N e atom%15Nbackground)/100 * Nconc * sample weight
with Nconc as the nitrogen concentration in mmol g1dry weight. Samples for Li analyses were microwave-digested in acid (0.1 g ground plant material, 1.5 ml H2O dest., 3 ml HNO3 (65%) and 1.5 ml H2O2 (30%); Mars Xpress, CEM GmbH, Kamp-Lintfort, Germany) and analysed with an atomic absorption spectrometer (AAS 5 EA, Analytik Jena Ag, Jena, Germany). Li concentrations (mmol g1dry weight) were multiplied with the sample weight to obtain Liexcess (mmol g1sample). All mass loss and tracer results were standardized to allow comparability among the different litter species in the analyses. For mass loss and the tracer and N release of labelled leaf litter, we calculated the remaining percent of biomass or tracer and N content at the respective timestep. For the tracer transfer to unlabelled leaf litter or tracer uptake by phytometer plants, we calculated the percentage of incorporated tracer from the initial tracer content of the corresponding labelled litter species. The results refer to the amount of incorporated tracer in the unlabelled litter species. Here, we further had to account for differences in the initial amounts of unlabelled leaf litter per species, which was unequal in 2- and 3species mixtures (2.55 and 1.7 g respectively), affecting the total amount of species specific tracer uptake by unlabelled litter. We thus counted only 2/3 of the uptake by litter in 2-species mixtures to adjust for its higher litter mass.
2.4. Tracer analysis After 12, 37, 101, and 171 days, litter and the phytometer plant of one randomly assigned mesocosm per plot were collected. Leaves and litter fragments were cleaned, sorted to species, dried at 40 C, and weighed. Litter fragments smaller than 5 mm in diameter could not be sorted to species; these fragments were combined to an “unsorted” sample that was weighed separately but not used for nutrient analyses. The share of each species in this unsorted sample was approximated, i.e. the litter pieces were evenly distributed to a continuous surface and the area percentage of each species was estimated by adapting the cover-abundance scale of Dierschke (1994). The leaf litter that was sorted to species was ground to
Table 1 Mean initial litter tracer and N concentrations in mmol g1dry Species
N litter
15
Q. acutissima S. saponaria S. superba
1143.7 1460.1 1350.7
7.8 11.4 14.9
N labelled litter
weight 15
We tested for effects of litter species identity, richness (nonlinear) and composition for all time-steps (12, 37, 101 and 171 days of decomposition) (i) on decomposition rates of mixed litters and of species specific litter, (ii) on the remaining tracer and N content in the litter mixtures, (iii) on tracer transfer among species, and (iv) on the uptake by phytometer plants. We used type I ANOVA (analysis of variance) models. For plot specific analyses across all species for each timestep, litter species identity (species), richness (species richness), and composition (composition) were included in the
and polyphenol concentrations in %
N unlabelled litter
4.0 4.6 4.4
2.5. Statistical analysis
dry weight
(polyphenol data were derived from Ristok et al., 2017).
Li labelled litter
Li unlabelled litter
Phenolics
Tannins
0.4 3.3 13.3
0.1 0.1 0.4
5e9 0.9e1 8e10
3e5 1 3e4
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
47
Table 2 Summarized results of ANOVA significance tests for plot and litter species specific litter mass loss, N and Li release at the specific time steps. (L) refers to results of labelled, (U) to results of unlabelled leaves. Horizontal lines indicate factors which could be not tested in the respective model or according to data specific circumstances, indicated by 1 (Q. acutissima was excluded from analyses due to very low initial Li contents of labelled litter). Significance levels: P < 0.001 (***), P < 0.01 (**), P < 0.05 (*), P < 0.1 (.). Qu (Q. acutissima), Sa (S. saponaria), Sc (S. superba). 15
Days of decomposition
Factors tested
Litter mass loss (L þ U) plot
Qu
Sa
Sc
plot
Qu
Sa
Sc
plot
Qu
Sa
Sc
plot
Qu
Sa
Sc
12
Species Species richness Composition
***
e
e
e
***
e
e
e
***
e
e
e
**
e e1 e1
e
e
37
Species Species richness Composition
***
e
e
e
***
* *** **
e e1 e1
e **
e .
101
Species Species richness Composition
***
e
e
e .
**
e *
**
*
e e1 e1
e
**
Species Species richness Composition
*** **
e
e .
*
e e1 e1
e
e
171
Litter
N release (L)
Litter N release (L)
**
Litter Li release (L)
* e
e .
e
e
e
.
e ***
e
e .
e ** .
.
e
***
e
.
*
**
e
e
.
e *
e
e
e
e
e
*
Table 3 Summarized results of ANOVA significance tests for plot and litter species specific tracer transfer among litter species (tracer uptake of unlabelled litter species from labelled litter species in %) and the 15N uptake by phytometer plants (tracer uptake of phytometers from labelled litter in %) at the specific time steps. (U) refers to results of unlabelled leaves. Horizontal lines indicate factors which could be not tested in the respective model or according to data specific circumstances, indicated by 1 (Q. acutissima was excluded from analyses due to very low initial Li contents of labelled litter) and 2 (in addition to 1, no Li uptake found for S. superba leaf litter). Significance levels: P < 0.001 (***), P < 0.01 (**), P < 0.05 (*), P < 0.1 (.). Qu (Q. acutissima), Sa (S. saponaria), Sc (S. superba). Days of decomposition
Factors tested
Litter 15N transfer among species (U)
Litter Li transfer among species (U)
15
plot
Qu
Sa
Sc
plot
12
Species Species richness Composition
***
e
e
e
37
101
171
.
Species Species richness Composition
e
Species Species richness Composition
e
Species Species richness Composition
***
e
e
e
e
model as fixed effects, and block and plot as random effects (Tables 2 and 3). For species specific analyses, the same model was applied without litter species identity. All data were analysed using R 2.13.0 (http://r-project.org) with the packages 'lme4and 'stats. 3. Results 3.1. Effects of litter species identity Litter species identity affected decomposition dynamics across all steps of mass loss, nutrient release (Table 2, Fig. 2, Tables S1eS4), transfer among litter species (Table 3, Fig. 3, Tables S5eS6) and uptake by phytometer plants (Table 3, Fig. 4, Table S7). Litter decomposition rates differed strongly between species (Table 2, P < 0.001 at all individual dates), with highest decomposition rates of S. saponaria leaf litter (k ¼ 8.33 ± 0.47 yr1, corresponding to a mean weight loss after 171 days of 98%), followed by S. superba (3.59 ± 0.28 yr1, 81% mass loss) and Q. acutissima (3.56 ± 0.24 yr1, 81% mass loss). This faster mass loss pattern of S. saponaria was also found for N and 15N loss (Fig. 2), and the 15N uptake by indicator plants (Fig. 4), while Li loss was similar for S. saponaria and S. superba. In contrast, Q. acutissima showed no Li release over time
Qu
Sa
Sc
plot
Qu
Sa
Sc
e
e
e e2 e2
***
e
e
e
e e2 e2
***
e
e
e
e e2 e2
.
e
e
e
e e2 e2
.
e
e
e1
*
.
e
** . ***
e * **
e1
** . ***
e
e
***
e
e
**
*
e1
e
e
N uptake by phytometer plants
e
e
1
. e
which was due to very low initial Li concentrations, leading to the decision to exclude it as a donor species in the statistical analyses of Li transfer among litter species (Table 3, Fig. 2). We observed species specific difference for the 15N and Li uptake of unlabelled litter species from labelled litter which were time-dependent (P < 0.001 at day 12 and 171 of decomposition for 15N and P < 0.01 for day 37 and 101 and P < 0.001 at day 171 for Li). Here, highest nutrient uptake was observed for Q. acutissima, followed by S. saponaria, while no uptake was found for S. superba (Fig. 3). 3.2. Effects of litter species diversity Litter species richness affected mass loss, 15N and Li release (Fig. 5) and Li transfer dynamics (Fig. 6). However, these effects were highly species specific, time-dependent and did not propagate across different processes (Tables 2 and 3). Species richness effects on nutrient transfer among litter species were different from those on mass loss and nutrient release. Litter mass loss in three species mixtures was slower compared to other species richness levels at day 171 of decomposition (P < 0.01; Table 2, Fig. 5). This effect was caused by negative species richness effects on S. saponaria (P < 0.001) and S. superba (P < 0.1) at this sampling
48
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
Fig. 1. Ratio of labelled and unlabelled leaf litter within monocultures, 2- and 3-species mixtures.
Fig. 2. Species specific mass loss and nutrient release of decomposing litter over time, with percent of remaining litter mass, 15N, N and Li after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors.
date. Release of 15N or N was never affected by litter species richness at the plot level. At the species level only 15N release of S. saponaria at day 171 was negatively affected (P < 0.01, Table 2), similar to the pattern found for mass loss. Litter Li release was faster in three species mixtures after 37 (P < 0.01) and 101 days
(P < 0.001) of decomposition due to positive effects of species richness in S. saponaria (P < 0.01) and S. superba (P < 0.1) at day 37 and S. superba (P < 0.05) at day 101 (Table 2). For Li transfer from labelled to unlabelled litter species, more Li was transferred in 2species mixtures at day 37 and 101 of decomposition (P < 0.1),
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
49
Fig. 3. Nutrient transfer among decomposing litter species, shown as percent of 15N and Li uptake of unlabelled litter from the corresponding labelled litter species after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). The nutrient uptake of S. superba is not displayed since no tracer uptake was observed. Displayed are mean values with standard errors. Q. acutissima was excluded as a donor species for Li transfer.
Fig. 4. 15N uptake by phytometer plants from labelled litter species after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors.
partly caused by higher uptake of Q. acutissima at day 37, while no effects were found for 15N transfer (Table 3; Fig. 6). Litter species composition affected mass loss (P < 0.01; day 101) and Li release (P < 0.01; day 37) but strongly depended on decomposition time and litter species identity (Table 2). We observed a consistent higher transfer of Li from labelled S. saponaria litter to unlabelled Q. acutissima leaves across all timesteps, while the 15N transfer to unlabelled litter of Q. acutissima (P < 0.1) and S. superba (P < 0.05) after 12 days was solely slightly affected by composition (Table 3). The 15N uptake by phytometer plants was neither affected by litter species richness nor by composition (Fig. 7, Table 3). 4. Discussion For the first time we traced two essential nutrients for plants though different processes of litter decomposition dynamics along a litter species richness gradient in a field trial. Our results provide new insights into the role of litter species identity and richness on nutrient dynamics in litter mixtures and the subsequent uptake by
plants. 4.1. Effects of litter species identity It is generally accepted that litter decomposition rates are highly species-specific. Litter quality traits (chemical composition and stoichiometry), determine the attractiveness of the food source for decomposer organisms, while leaf shape and toughness affect microhabitat conditions and physical access for decomposers (e.g. €ttenschwiler et al., reviewed in Gartner and Cardon, 2004; Ha 2005a). Species identity effects were also predominant in our study and propagated through all of the investigated parameters characterizing decomposition and nutrient dynamics. For example, litter of S. saponaria showed highest decomposition rates compared to the other species. This pattern was also observed for 15N release and uptake by plants from labelled leaves of this litter species. Leaves of S. saponaria were soft, with the lowest tensile strength of all species (Eichenberg D., personal communication). This speciesspecific leaf toughness has been recognized as an important
50
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
Fig. 5. Mass loss and nutrient release of decomposing litter over time, with percent of remaining litter mass, 15N, N and Li in monocultures, 2- and 3-species mixtures after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors.
Fig. 6. Nutrient transfer among decomposing litter species, shown as percent of 15N and Li transfer of unlabelled litter from the corresponding labelled litter species in 2-, and 3species mixtures after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors. Q. acutissima was excluded as a donor species for Li transfer.
predictor for litter decomposition rates (e.g. Cornelissen et al., 1999; rez-Harguindeguy et al., 2000), as it represents physical barriers Pe (e.g. cuticula, cell walls). Such barriers likely also explain part of species specific effects on nutrient transfer to unlabelled litter. For
example, S. superba showed no signs of nutrient transfer from other species (Fig. 3), which might be due to the leathery leaves with highest tensile strength of all three species (Eichenberg D., personal communication) that could negatively affect colonization by
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
51
Fig. 7. 15N uptake by phytometer plants from labelled monoculture, 2- and 3-species leaf litter after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors.
microorganisms such as fungi or bacteria. This assumption is supported by results on polyphenols of the litter used in our experiment reported by Ristok et al. (2017). Highest polyphenol contents were here found in S. superba, while lowest contents were observes for S. saponaria, which fits well to our data (Table 1). Also nutrient release and transfer dynamics showed different species specific patterns for 15N and Li, with a faster cation release of labelled and higher cation uptake of unlabelled leaves compared to 15N. In contrast to N, Li release was much faster and similar for S. saponaria and S. superba. These results suggest that release dynamics depend on the nutrient type, which is related to the chemical nature of these elements. For example, N is mainly chemically bound in organic molecules, while K and its chemically comparable tracer surrogate Li (Gockele et al., 2014) appear in highly soluble forms, resulting in a fast initial K and slower N release (e.g. Blair, 1988; Briones and Ineson, 1996; Yang et al., 2004; Schreeg et al., 2013). Further, initial nutrient amounts in litter and their stoichiometry affect the attractiveness of this litter species as a food source, which may explain species specific effects. Similar to our results, also Ristok et al. (2017) observed species identity effects on polyphenol composition (only four shared compounds found in S. superba and Q. acutissima) and on phenol and tannin decomposition rates. But also data of green leaves of the same species but another experiment showed species specific contents also for other nutrients (Table S8). These differences likely become less important when the specific nutrient is released very fast. All these results thus support our first hypothesis, in which we expected propagating species identity effects and nutrient specific differences in these highly interconnected steps of litter decomposition processes.
4.2. Effects of litter species richness In contrast to our second hypothesis, we did not observe general positive litter species richness effects on litter decomposition dynamics, and those few effects found were idiosyncratic, species specific and time-dependent. Further, these effects did not propagate through the studied processes of decomposition, which is also contrary to our expectation. However, species diversity mainly influenced Li dynamics, indicating that effects of litter species richness and composition on decomposition processes can be nutrient specific. Litter diversity effects on Li transfer from labelled litter to unlabelled Q. acutissima litter showed no effects on other decomposition processes of this species. These findings suggest
that the amount of nutrients transferred was too low to affect other decomposition processes or that species identity effects may outweigh diversity effects on nutrient transfer. Nutrient transfer among species in decomposing litter mixtures may thus be less important for litter diversity effects on decomposition processes compared to species identity effects. This is also in contrast to our third hypothesis in which we expected that observed richness effects on decomposition dynamics will be also found for the nutrient transfer among litter species. The few observed time and nutrient dependent diversity effects could have been caused by a range of factors. Especially the initial species-specific chemical composition (H€ attenschwiler et al., 2005b) and physical litter traits may affect a range of interconnected mechanisms during decomposition. For example, the physical structure of the litter layer determines the aggregation and hence litter contact areas within mixtures. Large contact areas between leaves likely enhance nutrient release but also transfer rates among different litter species. These contact areas very likely differ with species diversity and change with decomposition time. A dense packing, which usually increases with time, as well as litter traits such as toughness, but also particle size, can further positively affect water-holding capacities (e.g. Naeth et al., 1991; Dirks et al., 2010) and thus litter decomposition (Jin et al., 2013; Makkonen et al., 2013). We thus suggest that an increased water-holding capacity also positively affected cation release. Leaching of soluble cations would then be more sensitive towards litter packing and structural changes in mixtures compared to less soluble dissolved organic N (Hagedorn and Machwitz, 2007). This assumption is supported by our results, which show litter richness effects on Li but not on 15N contents. In contrast to Li, mass loss and 15N release showed time-dependent species diversity effects, with slower mass loss and 15N release in 3-species mixtures at day 171. We do not know if and to which extent this is an artefact, e.g. due to random variation, too low values at the third time step (no increase at the last time step), due to contamination of smaller litter particles with mineral material or by a stronger fungal colonization in diverse mixtures at the last time step. However, it is also possible that such negative effects were caused by inhibiting substances of species which hamper the attractiveness to decomposer of other species in mixtures (Schimel et al., 1998). If one species would release such inhibiting substances, the higher probability to have that species present in mixtures would thus explain an overall negative diversity effect (sampling effect, Tilman et al., 2014). It is likely that these two different
52
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
mechanisms for highly soluble and moderate soluble nutrients, respectively, were operating in our study system: Water holding capacity might have been important for leaching of the soluble cation, while mass loss and N release were likely hampered by retardants. This is supported by other studies, showing that decomposer organisms are not affected by water holding capacity (e.g. Taylor et al., 2004; Bruder et al., 2011) but by inhibiting substances (Schimel et al., 1998). Although species richness slowed litter and 15N mass loss in three species mixtures after 171 days of decomposition, species richness did not affect the 15N uptake by phytometer plants. This suggests that only strong differences among litter species in mass loss may affect short term uptake rates by plants. Slight diversity effects on mass loss and nutrient release are thus likely buffered by the soil (i.e. nutrients from decomposing litter contribute to an already existent nutrient pool in the soil which likely buffers temporal differences in nutrient input) and need more time and decomposed litter of the respective litter mixtures to develop. Species richness effects on litter decomposition would then not affect nutrient uptake dynamics by plants after a short time span of 171 days. Diversity effects on different ecosystem processes often need time to develop (e.g. Fargione et al., 2007; Reich et al., 2012) while species identity effects occur from the beginning and remain important over time (e.g. Vivanco and Austin, 2008). We observed similar effects in our decomposition experiment for litter mass loss over time. Here, species identity affected the whole decomposition process, while richness effects on mass loss occurred solely at the end of the experiment. Several interconnected processes such as decomposer colonization and interactions among decomposer guilds, and the transfer and release of nutrients or inhibiting substances from in litter mixtures need time to develop and likely increase over time with litter packing. 5. Conclusion We conclude that the use of tracer elements is a powerful technique to investigate the different processes of nutrient dynamics within decomposing litter, ranging from mass loss and associated nutrient release and transfer among litter species, to the subsequent uptake of the mineralized nutrients by plants. We were thus able to demonstrate for the first time the potential impacts of litter species diversity and identity for these different processes involved in decomposition of organic matter. Particularly, our results confirm the dominant effect of species identity and composition - and thus the functional importance of plant traits - not only on mass loss, but also on related nutrient dynamics. In addition, species richness seems to be relevant only at certain points in time for nutrient dynamics in decomposing litter, but not necessarily for the final uptake of N by plants. This latter result supports the prior assumption that litter species mixture effects on decomposition rates may have no direct implications on plant nutrition (e.g. Wardle et al., 1997), at least in subtropical climate zones and for the observed litter species. Plant nutrition is thus rather determined by species specific nutrient contents which are released during decomposition. Nevertheless, our study reinforces the concept of strong biotic control over decomposition as a key process in ecosystem functioning, which opens the possibility to actively use the huge diversity of different tree species in management of ecosystem services provided by forests. Acknowledgements We thank our cooperation partners Zhiqin Pei, Katherina A. Pietsch, David Eichenberg, Junfei Guo, Jessica Gutknecht, Sabine
Both, Manuel Hib, Helge Bruelheide and many other people for their help to establish the experimental setup. We thank Steffen Seitz, Phillip Goebes from the University of Tübingen for soil analysis and Erika Fischer from the Centre for Biological Systems Analysis at the University of Freiburg for IRMs analysis. We are grateful for the help of Ulrike Erhard, Benjamin Plaga, Wolfgang H. Müller and student helpers who made our intensive lab work possible and Angela L. de Avila for comments on a former version of this manuscript. We further thank Barbara Weisbrod and Benjamin Plaga for their help with collecting data presented in Table S8. We are grateful for the funding provided by the German Research Foundation (DFG FOR 891/2). We further thank the Sino-German Centre for Research Promotion for funding a Summer School in Jingdezhen in 2015 (GZ 1146) about scientific writing, where parts of this article were written. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.soilbio.2017.08.007. References Ball, B.A., Hunter, M.D., Kominoski, J.S., Swan, C.M., Bradford, M.A., 2008. Consequences of non-random species loss for decomposition dynamics: experimental evidence for additive and non-additive effects. Journal of Ecology 96, 303e313. http://dx.doi.org/10.1111/j.1365-2745.2007.01346.x. Berg, B., Berg, M.P., Bottner, P., Box, E., Breymeyer, A., Anta, R.C., de, Couteaux, M., €lko € nen, E., Escudero, A., Gallardo, A., Kratz, W., Madeira, M., Ma ~ oz, F., Piussi, P., Remacle, J., Santo, A.V. McClaugherty, C., Meentemeyer, V., Mun de, 1992. Litter mass loss rates in pine forests of Europe and Eastern United States: some relationships with climate and litter quality. Biogeochemistry 20, 127e159. http://dx.doi.org/10.1007/BF00000785. Berg, B., McClaugherty, C., 2008. Plant Litter: Decomposition, Humus Formation, Carbon Sequestration, 2nd. ed. Springer, Berlin. Blair, J.M., 1988. Nutrient release from decomposing foliar litter of three tree species with spicial reference to calcium, magnesium and potassium dynamics. Plant and Soil 110, 49e55. Briones, M.J.I., Ineson, P., 1996. Decomposition of eucalyptus leaves in litter mixtures. Soil Biology and Biochemistry 28, 1381e1388. http://dx.doi.org/10.1016/ S0038-0717(96)00158-7. Bruder, A., Chauvet, E., Gessner, M.O., 2011. Litter diversity, fungal decomposers and litter decomposition under simulated stream intermittency: litter diversity, stream intermittency and decomposition. Functional Ecology 25, 1269e1277. http://dx.doi.org/10.1111/j.1365-2435.2011.01903.x. Caner, L., Zeller, B., Dambrine, E., Ponge, J.-F., Chauvat, M., Llanque, C., 2004. Origin of the nitrogen assimilated by soil fauna living in decomposing beech litter. Soil Biology and Biochemistry 36, 1861e1872. Cardinale, B.J., Duffy, J.E., Gonzalez, A., Hooper, D.U., Perrings, C., Venail, P., Narwani, A., Mace, G.M., Tilman, D., Wardle, D.A., Kinzig, A.P., Daily, G.C., Loreau, M., Grace, J.B., Larigauderie, A., Srivastava, D.S., Naeem, S., 2012. Biodiversity loss and its impact on humanity. Nature 486, 59e67. http:// dx.doi.org/10.1038/nature11148. Cardisch, G., Giller, K.E., 1996. Driven by Nature: Plant Litter Quality and Decomposition, first ed. CABI, Wallingford, Oxon, UK. Chapman, K., Whittaker, J.B., Heal, O.W., 1988. Proceedings of a workshop on interactions between soil-inhabiting invertebrates and microorganisms in relation to plant growth metabolic and faunal activity in litters of tree mixtures compared with pure stands. Agriculture, Ecosystems & Environment 24, 33e40. http://dx.doi.org/10.1016/0167-8809(88)90054-0. rez-Harguindeguy, N., Díaz, S., Grime, J.P., Marzano, B., Cornelissen, J.H., Pe Cabido, M., Vendramini, F., Cerabolini, B., 1999. Leaf structure and defence control litter decomposition rate across species and life forms in regional floras on two continents. New Phytologist 143, 191e200. Dierschke, H., 1994. Pflanzensoziologie: Grundlagen und Methoden, first ed. UTB, Stuttgart, Stuttgart. Dirks, I., Navon, Y., Kanas, D., Dumbur, R., Grünzweig, J.M., 2010. Atmospheric water vapor as driver of litter decomposition in Mediterranean shrubland and grassland during rainless seasons. Global Change Biology 16, 2799e2812. http://dx.doi.org/10.1111/j.1365-2486.2010.02172.x. Fahey T.J., Yavitt, J.B., Sherman, R.E., Groffman, P.M., Fisk, M.C., Maerz, J.C., 2011. Transport of carbon and nitrogen between litter and soil organic matter in a Northern hardwood forest. Ecosystems 14, 326e340. http://dx.doi.org/10.1007/ s10021-011-9414-1. Fargione, J., Tilman, D., Dybzinski, R., Lambers, J.H.R., Clark, C., Harpole, W.S., Knops, J.M.H., Reich, P.B., Loreau, M., 2007. From selection to complementarity: shifts in the causes of biodiversityeproductivity relationships in a long-term biodiversity experiment. Proceedings of the Royal Society of London B:
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53 Biological Sciences 274, 871e876. http://dx.doi.org/10.1098/rspb.2006.0351. Fitter, A.H., 1986. Spatial and temporal patterns of root activity in a species-rich alluvial grassland. Oecologia 69, 594e599. http://dx.doi.org/10.1007/ BF00410368. Frey, S.D., Six, J., Elliott, E.T., 2003. Reciprocal transfer of carbon and nitrogen by decomposer fungi at the soilelitter interface. Soil Biology and Biochemistry 35, 1001e1004. http://dx.doi.org/10.1016/S0038-0717(03)00155-X. Fyles, J.W., Fyles, I.H., 1993. Interaction of Douglas-fir with red alder and salal foliage litter during decomposition. Canadian Journal of Forest Research 23, 358e361. http://dx.doi.org/10.1139/x93-052. Gartner, T.B., Cardon, Z.G., 2004. Decomposition dynamics in mixed-species leaf litter. Oikos 104, 230e246. http://dx.doi.org/10.1111/j.0030-1299.2004.12738.x. Gockele, A., Weigelt, A., Gessler, A., Scherer-Lorenzen, M., 2014. Quantifying resource use complementarity in grassland species: a comparison of different nutrient tracers. Pedobiologia 57, 251e256. http://dx.doi.org/10.1016/ j.pedobi.2014.09.001. Hagedorn, F., Machwitz, M., 2007. Controls on dissolved organic matter leaching from forest litter grown under elevated atmospheric. Soil Biology and Biochemistry 39, 1759e1769. http://dx.doi.org/10.1016/j.soilbio.2007.01.038. €ttenschwiler, S., Tiunov, A.V., Scheu, S., 2005a. Biodiversity and litter decompoHa sition in terrestrial ecosystems. Annual Review of Ecology, Evolution, and Systematics 36, 191e218. http://dx.doi.org/10.1146/ annurev.ecolsys.36.112904.151932. €ttenschwiler, S., Tiunov, A.V., Scheu, S., 2005b. Biodiversity and litter decompoHa sition in terrestrial ecosystems. Annual Review of Ecology, Evolution, and Systematics 36, 191e218. http://dx.doi.org/10.1146/ annurev.ecolsys.36.112904.151932. Hooper, D.U., Chapin, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H., €, H., Symstad, A.J., Lodge, D.M., Loreau, M., Naeem, S., Schmid, B., Set€ ala Vandermeer, J., Wardle, D.A., 2005. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecological Monographs 75, 3e35. http://dx.doi.org/10.1890/04-0922. IUSS Working Group WRB, 2015. World Reference Base for Soil Resources 2014, Update 2015. International soil classification system for naming soils and creating legends for soil maps. Jin, V.L., Haney, R.L., Fay, P.A., Polley, H.W., 2013. Soil type and moisture regime control microbial C and N mineralization in grassland soils more than atmospheric CO 2-induced changes in litter quality. Soil Biology and Biochemistry 58, 172e180. Johnson, D.W., Todd, D.E., 1998. Effects of Harvesting Intensity on Forest Productivity and Soil Carbon Storage in a Mixed Oak Forest. Kaneko, N., Salamanca, E., 1999. Mixed leaf litter effects on decomposition rates and soil microarthropod communities in an oakepine stand in Japan. Ecological Research 14, 131e138. http://dx.doi.org/10.1046/j.1440-1703.1999.00292.x. Lummer, D., Scheu, S., Butenschoen, O., 2012. Connecting litter quality, microbial community and nitrogen transfer mechanisms in decomposing litter mixtures. Oikos 121, 1649e1655. http://dx.doi.org/10.1111/j.1600-0706.2011.20073.x. Makkonen, M., Berg, M.P., van Logtestijn, R.S.P., van Hal, J.R., Aerts, R., 2013. Do physical plant litter traits explain non-additivity in litter mixtures? A test of the improved microenvironmental conditions theory. Oikos 122, 987e997. http:// dx.doi.org/10.1111/j.1600-0706.2012.20750.x. McArthur, J. m, 1994. Recent trends in strontium isotope stratigraphy. Terra Nova 6, 331e358. http://dx.doi.org/10.1111/j.1365-3121.1994.tb00507.x. McTiernan, K.B., Ineson, P., Coward, P.A., 1997. Respiration and nutrient release from tree leaf litter mixtures. Oikos 78, 527. http://dx.doi.org/10.2307/3545614. Naeth, M.A., Bailey, A.W., Chanasyk, D.S., Pluth, D.J., 1991. Water holding capacity of litter and soil organic matter in mixed prairie and fescue grassland ecosystems of Alberta. Journal of Range Management 44, 13e17. http://dx.doi.org/10.2307/ 4002630. rez-Harguindeguy, N., Díaz, S., Cornelissen, J.H., Vendramini, F., Cabido, M., Pe
53
Castellanos, A., 2000. Chemistry and toughness predict leaf litter decomposition rates over a wide spectrum of functional types and taxa in central Argentina. Plant and Soil 218, 21e30. http://dx.doi.org/10.1023/A:1014981715532. Peterson, D.L., Rolfe, G.L., 1982. Nutrient dynamics and decomposition of litterfall in floodplain and upland forests of central illinois. Forest Science 28, 667e681. Reich, P.B., Tilman, D., Isbell, F., Mueller, K., Hobbie, S.E., Flynn, D.F.B., Eisenhauer, N., 2012. Impacts of biodiversity loss escalate through time as redundancy fades. Science 336, 589e592. http://dx.doi.org/10.1126/science.1217909. Ristok, C., Leppert, K.N., Franke, K., Scherer-Lorenzen, M., Niklaus, P.A., Wessjohann, L.A., Bruelheide, H., 2017. Leaf litter diversity positively affects the decomposition of plant polyphenols. Plant and Soil 416, 1e13. http://dx.doi.org/ 10.1007/s11104-017-3340-8. 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. Biogeochemistry 42, 221e234. http://dx.doi.org/10.1023/A: 1005911118982. €ttenschwiler, S., 2007. Nitrogen transfer between decomposing Schimel, J.P., Ha leaves of different N status. Soil Biology and Biochemistry 39, 1428e1436. http://dx.doi.org/10.1016/j.soilbio.2006.12.037. Schreeg, L.A., Mack, M.C., Turner, B.L., 2013. Nutrient-specific solubility patterns of leaf litter across 41 lowland tropical woody species. Ecology 94, 94e105. Seastedt, T.R., 1984. The role of microarthropods in decomposition and mineralization processes. Annual Review of Entomology 29, 25e46. http://dx.doi.org/ 10.1146/annurev.en.29.010184.000325. Seitz, S., Goebes, P., Zumstein, P., Assmann, T., Kühn, P., Niklaus, P.A., Schuldt, A., Scholten, T., 2015. The influence of leaf litter diversity and soil fauna on initial soil erosion in subtropical forests. Earth Surface Processes and Landforms 40, 1439e1447. http://dx.doi.org/10.1002/esp.3726. € ter, D., Pflug, A., Wolters, V., 2004. Response of different decomTaylor, A.R., Schro poser communities to the manipulation of moisture availability: potential effects of changing precipitation patterns. Global Change Biology 10, 1313e1324. http://dx.doi.org/10.1111/j.1365-2486.2004.00801.x. Tilman, D., Isbell, F., Cowles, J.M., 2014. Biodiversity and ecosystem functioning. Annual Review of Ecology, Evolution, and Systematics 45, 471e493. http:// dx.doi.org/10.1146/annurev-ecolsys-120213-091917. Tiunov, A.V., 2009. Particle size alters litter diversity effects on decomposition. Soil Biology and Biochemistry 41, 176e178. http://dx.doi.org/10.1016/ j.soilbio.2008.09.017. Vivanco, L., Austin, A.T., 2008. Tree species identity alters forest litter decomposition through long-term plant and soil interactions in Patagonia, Argentina. Journal of Ecology 96, 727e736. http://dx.doi.org/10.1111/j.1365-2745.2008.01393.x. Wardle, D.A., Bonner, K.I., Nicholson, K.S., 1997. Biodiversity and plant litter: experimental evidence which does not support the view that enhanced species richness improves ecosystem function. Oikos 79, 247. http://dx.doi.org/10.2307/ 3546010. €ber, W., Ma, K., Nadrowski, K., Yang, X., Bauhus, J., Both, S., Fang, T., H€ ardtle, W., Kro Pei, K., Scherer-Lorenzen, M., Scholten, T., Seidler, G., Schmid, B., von Oheimb, G., Bruelheide, H., 2013. Establishment success in a forest biodiversity and ecosystem functioning experiment in subtropical China (BEF-China). European Journal of Forest Research 132, 593e606. http://dx.doi.org/10.1007/s10342-0130696-z. Yang, Y.S., Guo, J.F., Chen, G.S., Xie, J.S., Cai, L.P., Lin, P., 2004. Litterfall, nutrient return, and leaf-litter decomposition in four plantations compared with a natural forest in subtropical China. Annals of Forest Science 61, 465e476. Zeller, B., Colin-Belgrand, M., Dambrine, E., Martin, F., 2001. Fate of nitrogen released from 15N-labeled litter in European beech forests. Tree Physiology 21, 153e162. Zeller, B., Colin-Belgrand, M., Dambrine, E., Martin, F., Bottner, P., 2000. Decomposition of 15N-labelled beech litter and fate of nitrogen derived from litter in a beech forest. Oecologia 123, 550e559. http://dx.doi.org/10.1007/PL00008860.
Contents lists available at ScienceDirect
Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio
Does species richness of subtropical tree leaf litter affect decomposition, nutrient release, transfer and subsequent uptake by plants? Katrin N. Leppert a, *, Pascal A. Niklaus b, Michael Scherer-Lorenzen a a b
University of Freiburg, Faculty of Biology, Geobotany, Schaenzlestr. 1, 79104 Freiburg, Germany University of Zurich, Institute of Evolutionary Biology and Environmental Studies, Winterthurerstrasse 190, 8057 Zurich, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 27 March 2017 Received in revised form 18 July 2017 Accepted 3 August 2017
During leaf litter decomposition, nutrients are released, can be transferred among different litter species, are metabolized by soil organisms and are taken up by plants again. However, it remains unclear to which extent leaf litter species richness affects these processes of nutrient cycling, and whether effects on one of those processes propagate to the subsequent one. We established a common garden decomposition experiment in a Chinese subtropical secondary forest, to trace two essential nutrients during decomposition and their uptake by plants along a litter species diversity gradient. Unlabelled, and 15N and Li (as surrogate for K) labelled leaves of three native tree species were used to create replicated 1-, 2and 3-species mixtures, each with one species labelled per mixture. Litter mixtures were placed in mesocosms with one growing herbaceous phytometer plant. Over six months, litter and phytometer plants of each mixture were sampled at four points in time and the different process steps of nutrient dynamics were determined. Our results showed species and nutrient specific decomposition dynamics, which propagated through the processes of mass loss, nutrient release and transfer among species, and nutrient uptake dynamics of phytometer plants. However, we found no litter species diversity effects along the different litter decomposition processes. Rather specific diversity effects occurred in few cases at different points in time for mass loss, Li release and transfer dynamics. These effects were not caused by nutrient transfer from labelled to unlabelled litter, suggesting that species identity effects on decomposer dynamics may outweigh effects of nutrient transfer among litter species in mixtures. Further, the observed litter species diversity effects did not affect the 15N uptake of phytometer plants. Hence, the influence of species diversity on nutrient cycling and plant available nutrient stocks is mainly determined by the amount and variety of chemical compounds that different species exhibit and release to the soil. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Litter species diversity Leaf litter decomposition Nutrient dynamics Tracer elements BEF-China Subtropical forests
1. Introduction The significance of plant biodiversity for terrestrial ecosystem functioning and the delivery of ecosystem services has been pointed out over the last decades, emphasizing a strong biotic control over ecological processes (e.g. Hooper et al., 2005; Cardinale et al., 2012; Tilman et al., 2014). Initially focusing on the consequences of biodiversity loss on productivity, many studies now show that plant species diversity and functional composition of communities strongly affect a large variety of ecosystem properties
* Corresponding author. E-mail address: [email protected] (K.N. Leppert). http://dx.doi.org/10.1016/j.soilbio.2017.08.007 0038-0717/© 2017 Elsevier Ltd. All rights reserved.
and processes. Among these, decomposition of dead organic matter is a key function as it affects nutrient cycling, soil formation and carbon storage (Johnson and Todd, 1998; Berg and McClaugherty, 2008). A range of experiments has evaluated the effects of plant diversity on decomposition of leaf litter, demonstrating highly species-specific litter decomposition rates that depend on litter quality traits such as C:N or lignin:N ratios, secondary metabolites or morphological characteristics (e.g. Berg et al., 1992; Cardisch and rez-Harguindeguy et al., 2000). Decomposition rates Giller, 1996; Pe of litter species in mixtures often differ from the average ones in their component monocultures, i.e. they are non-additive (e.g. Seastedt, 1984; Chapman et al., 1988; Ball et al., 2008). Thus, some leaves decompose faster (synergistic effects) or slower
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
(antagonistic effects) than one would expect from respective monoculture decomposition rates or show idiosyncratic responses (non-linear richness effects) (Gartner and Cardon, 2004; €ttenschwiler et al., 2005a). Litter species richness effects can Ha be caused by (1) an increased attractiveness of low quality litter as a food source for decomposers if mixed with high quality litter, via nutrient transfer among leaves of different species, (2) variations of species-specific litter compounds with either stimulating or inhibiting properties, (3) alteration of microclimate and habitat structures through differences in leaf shapes and toughness, and (4) changes in the abundance or activity of certain decomposer guilds, and (5) feedback mechanisms among different decomposer guilds €ttenschwiler et al., 2005a). (reviewed by Ha Seeing litter decomposition simply as the rate of mass loss or CO2-production, as done in most studies, does not adequately reflect the underlying complex, time-dependent phenomena such as leaching of soluble compounds, physical breakup, and biotic shredding, enzymatic breakage of complex compounds, release and transfer of elements among different litter types through leaching and decomposers (e.g. reviewed in Cardisch and Giller, 1996; Berg and McClaugherty, 2008). Potential diversity effects on some of these individual processes, such as the nutrient transfer among litter species in mixtures or the subsequent nutrient uptake by plants, have been rarely considered and remain unclear. Nutrient concentrations of decomposing litter species in mixtures can increase or decrease compared to their respective monocultures (e.g. Briones and Ineson, 1996; McTiernan et al., 1997; Kaneko and Salamanca, 1999). One of the main potential mechanisms is the transport of nutrients among different litter species by leaching and diffusion (Fyles and Fyles, 1993; McArthur, 1994; Briones and Ineson, 1996) or transport by fungal hyphae (Frey et al., 2003; Tiunov, 2009; Lummer et al., 2012) which can improve decomposability of more recalcitrant litter species. Further, nutrients may partially originate from external sources (e.g. Peterson and Rolfe, 1982), such as fungal colonization (Caner et al., 2004) and fungal transport of deposited N from soil (Fahey et al., 2011). Whether subsequent richness effects on litter mass loss and nutrient release rates affect the nutrient uptake by plants still remains unclear. The study of such detailed decomposition processes and their underlying mechanisms is methodological challenging and ideally requires 1) a separation of the leaf litter according to species and 2) the use of isotope tracer elements to determine the nutrient sources and sinks. Common tracer elements to follow nutrients through the plant-soil system are stable isotopes like 15N, or rare elements which serve no apparent vital biological function but have similar chemical properties as other elements essential for plant growth. Lithium (Li), for example, can be used to quantify plant uptake of potassium (K) (Gockele et al., 2014) as they share Kþ transport carriers in roots (Fitter, 1986). By using tracers it is also possible to detect and quantify element transfer from labelled (donor) to €ttenschwiler, unlabelled (acceptor) leaf litter (e.g. Schimel and Ha 2007; Lummer et al., 2012). For example, Lummer et al. (2012) revealed nitrogen transfer through fungal hyphae from nitrogenpoor to nitrogen-rich leaf litter. These results challenge the traditional view of nitrogen transfer being one-directional from nitrogen-rich to nitrogen-poor litter species, and suggest that nutrient transfer mechanisms among litter species are not completely understood. Tracer elements have also been used to follow the transport of nutrients from decomposing leaves to soil and plants (e.g. Zeller et al., 2000, 2001) including the uptake by decomposer organisms (e.g. Caner et al., 2004). Zeller et al. (2000, 2001) documented a 1:1 relationship between litter mass loss and nitrogen release. After two years of litter decomposition, 35% of the nitrogen remained in the recalcitrant part of the litter, 50% reached the topsoil and about 2e4% was incorporated into 15e50
45
year old beech trees (Zeller et al., 2001). However, these studies focused on pure or low-diverse litter mixtures only, neglecting litter mixtures, and macro-nutrients other than nitrogen. Although these studies provide important insights into mechanisms of nutrient dynamics, they are thus only of limited use for the understanding of litter diversity effects on decomposition and nutrient cycling, especially within litter mixtures. Here, we specifically addressed the question of how litter species richness affects overall litter decomposition rates (i.e. mass loss), nutrient release and transfer among leaves in mixtures and the subsequent uptake by plants. We conducted a field litter decomposition experiment in the subtropics of south-east China and focused on the macro-nutrients N and K, which were represented by the tracers 15N and Li, to include one immobile and one highly mobile element in leaf litter (Briones and Ineson, 1996; Yang et al., 2004). 15N and Li labelled and unlabelled, senescent leaves of three native tree species were incubated in situ as monocultures, 2and 3-species mixtures in mesocosms and decomposition rates and tracer contents in the labelled and unlabelled litter species were followed over 24 weeks. We further used a phytometer approach to quantify nutrient transfer dynamics from leaves into plants. We tested three hypotheses: (i) Temporal dynamics for mass loss, nutrient release, transfer and uptake by plants are highly species- and nutrient specific. Due to strong functional links among these processes involved in decomposition, species identity effects will be consistent for these processes. (ii) Litter species richness positively affects litter mass loss and nutrient dynamics. These richness effects will propagate to subsequent processes, and will therefore also be found for nutrient release and uptake by phytometer plants. (iii) In case of litter species richness effects on litter mass loss, nutrient release and the uptake by indicator plants we expect richness effects also on nutrient transfer among litter species. This expectation is based on the fact that nutrient transfer among litter species has been hypothesized to be a key mechanism explaining diversity effects on decomposition processes €ttenschwiler et al., 2005a). (Ha 2. Materials and methods 2.1. Study site We conducted a litter decomposition experiment from Apr. (2012) to Oct. 2012 in Xingangshan, Jiangxi province in south-east China (29.1200 N, 117.91” E). The climate is subtropical with warm, wet summers and cold, dry winters. Mean annual temperature is 16.7 C and annual precipitation averages 1821 mm (Yang et al., 2013). The field site was situated on a mountain ridge in a small, natural mixed forest stand with Castanopsis calesii (HEMSLEY) HAYATA, Loropetalum chinense (R. BROWN) OLIVER, Eurya muricata DUNN, Lithocarpus glaber (THUNBERG) NAKAI and Castanopsis eyrei (CHAMPION EX BENTHAM) TUTCHER as dominant tree species (Eichenberg, personal communication). Besides few plants of Carex spec., no herb layer existed. The site is situated on an Endoleptic Cambisol (IUSS Working Group WRB, 2015) on silt loam (24% sand, 50% silt and 26% clay) with a mean pH in the topsoil layer (0e5 cm) of 4.7 (Seitz et al., 2015). 2.2. Production of labelled leaf litter We established a nursery with the native tree species Sapindus saponaria LINNAEUS, Quercus acutissima CARRUTHERS and Schima superba GARDNER & CHAMPION to produce both labelled and unlabelled leaf litter. A total of approx. 6000 3e6 year old trees of all three species were planted in spring 2011. One third of the trees was watered twice with 200 ml 0.001 M 15NH4Cl and 0.236 M LiCl,
46
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
respectively. Labelled and unlabelled trees were separated by natural borders or coarse meshed fences to prevent mixing of labelled and unlabelled leaf litter. Freshly fallen leaves were collected weekly from the ground in autumn 2011, omitting damaged ones. Leaves were pooled by species and label status (labelled and unlabelled), and dried at 40 C. Mean initial tracer concentrations in labelled leaf litter indicated a successful tracer uptake for almost all species (Table 1). Only mixtures with labelled Q. acutissima leaves were excluded from Li transfer statistics due to very low initial Li concentrations (Tables 2 and 3 comment ‘1’). 2.3. Experimental design and data collection We created a total of seven different litter species combinations, including three monocultures, three 2- and one 3-species mixture. Each litter mixture was realized with one species labelled per mixture, resulting in twelve mixtures with different label combinations which were replicated in three blocks (36 plots in total). Each mixture comprised 1/3 labelled and 2/3 unlabelled leaves to achieve similar total 15N and Li contents for each labelled species per mixture. For example, monocultures of S. superba contained 1/3 labelled and 2/3 unlabelled litter of this species; a 2-species mixture of labelled S. superba and unlabelled S. saponaria leaves contained 1/3 labelled and 1/6 unlabelled S. superba litter and 1/2 unlabelled S. saponaria litter; a 3-species mixture with labelled S. superba litter contained 1/3 labelled S. superba, 1/3 unlabelled S. saponaria and 1/3 unlabelled Q. acutissima litter (Fig. 1). These litter mixtures were placed in mesocosms made of PVC (15 cm diameter x 12.5 cm height, lowered 2 cm into the soil), allowing direct leaf-soil contact. Every mesocosm was filled with 5.1 g of the respective litter mixture. One phytometer plant (Carex sp.), originating from the same field site, was transplanted in the centre. The plants were excavated, dry leaves removed, and roots carefully cleaned with water prior to planting. Mesocosms were covered with a 1.5 cm mesh to prevent litter input from surrounding trees. Each plot contained four mesocosms (temporal replicates) of the same species mixture and covered an area of 45 45 cm. Overall, there were a total of 144 mesocosms. The position of all plots within the three blocks were chosen by randomization to account for a possible heterogeneity in soil conditions due to differences of the overstory species composition.
powder (Mixer Mill MM 200, Retsch GmbH, Haan, Germany) and analysed for 15N and Li contents. Total tracer pools were calculated by multiplying tracer concentrations with the total litter weight of the species, including the part estimated in the unsorted fraction. Each phytometer plant was harvested, freed from dead plant material at the base and analysed for its 15N content. Due to small sample volumes, Li could not be determined in the phytometers. First order litter decomposition rate constants k (yr1) for plot and species specific litter mass loss were calculated by fitting Xt ¼ X0 ekt to the data, where X0 and Xt are litter mass at time 0 and t, respectively. The fit of the function was inspected visually and generally matched the data well. 15 N was analysed with an isotope-ratio mass spectrometer (Thermo Fisher Scientific, Finnigan DELTAplus, Waltham, USA) at the Centre for Biological Systems Analysis in Freiburg. Total tracer 15 N contents (mmol sample1) are reported as 15Nexcess and were calculated with the formula 15 Nexcess ¼ (atom%15N e atom%15Nbackground)/100 * Nconc * sample weight
with Nconc as the nitrogen concentration in mmol g1dry weight. Samples for Li analyses were microwave-digested in acid (0.1 g ground plant material, 1.5 ml H2O dest., 3 ml HNO3 (65%) and 1.5 ml H2O2 (30%); Mars Xpress, CEM GmbH, Kamp-Lintfort, Germany) and analysed with an atomic absorption spectrometer (AAS 5 EA, Analytik Jena Ag, Jena, Germany). Li concentrations (mmol g1dry weight) were multiplied with the sample weight to obtain Liexcess (mmol g1sample). All mass loss and tracer results were standardized to allow comparability among the different litter species in the analyses. For mass loss and the tracer and N release of labelled leaf litter, we calculated the remaining percent of biomass or tracer and N content at the respective timestep. For the tracer transfer to unlabelled leaf litter or tracer uptake by phytometer plants, we calculated the percentage of incorporated tracer from the initial tracer content of the corresponding labelled litter species. The results refer to the amount of incorporated tracer in the unlabelled litter species. Here, we further had to account for differences in the initial amounts of unlabelled leaf litter per species, which was unequal in 2- and 3species mixtures (2.55 and 1.7 g respectively), affecting the total amount of species specific tracer uptake by unlabelled litter. We thus counted only 2/3 of the uptake by litter in 2-species mixtures to adjust for its higher litter mass.
2.4. Tracer analysis After 12, 37, 101, and 171 days, litter and the phytometer plant of one randomly assigned mesocosm per plot were collected. Leaves and litter fragments were cleaned, sorted to species, dried at 40 C, and weighed. Litter fragments smaller than 5 mm in diameter could not be sorted to species; these fragments were combined to an “unsorted” sample that was weighed separately but not used for nutrient analyses. The share of each species in this unsorted sample was approximated, i.e. the litter pieces were evenly distributed to a continuous surface and the area percentage of each species was estimated by adapting the cover-abundance scale of Dierschke (1994). The leaf litter that was sorted to species was ground to
Table 1 Mean initial litter tracer and N concentrations in mmol g1dry Species
N litter
15
Q. acutissima S. saponaria S. superba
1143.7 1460.1 1350.7
7.8 11.4 14.9
N labelled litter
weight 15
We tested for effects of litter species identity, richness (nonlinear) and composition for all time-steps (12, 37, 101 and 171 days of decomposition) (i) on decomposition rates of mixed litters and of species specific litter, (ii) on the remaining tracer and N content in the litter mixtures, (iii) on tracer transfer among species, and (iv) on the uptake by phytometer plants. We used type I ANOVA (analysis of variance) models. For plot specific analyses across all species for each timestep, litter species identity (species), richness (species richness), and composition (composition) were included in the
and polyphenol concentrations in %
N unlabelled litter
4.0 4.6 4.4
2.5. Statistical analysis
dry weight
(polyphenol data were derived from Ristok et al., 2017).
Li labelled litter
Li unlabelled litter
Phenolics
Tannins
0.4 3.3 13.3
0.1 0.1 0.4
5e9 0.9e1 8e10
3e5 1 3e4
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
47
Table 2 Summarized results of ANOVA significance tests for plot and litter species specific litter mass loss, N and Li release at the specific time steps. (L) refers to results of labelled, (U) to results of unlabelled leaves. Horizontal lines indicate factors which could be not tested in the respective model or according to data specific circumstances, indicated by 1 (Q. acutissima was excluded from analyses due to very low initial Li contents of labelled litter). Significance levels: P < 0.001 (***), P < 0.01 (**), P < 0.05 (*), P < 0.1 (.). Qu (Q. acutissima), Sa (S. saponaria), Sc (S. superba). 15
Days of decomposition
Factors tested
Litter mass loss (L þ U) plot
Qu
Sa
Sc
plot
Qu
Sa
Sc
plot
Qu
Sa
Sc
plot
Qu
Sa
Sc
12
Species Species richness Composition
***
e
e
e
***
e
e
e
***
e
e
e
**
e e1 e1
e
e
37
Species Species richness Composition
***
e
e
e
***
* *** **
e e1 e1
e **
e .
101
Species Species richness Composition
***
e
e
e .
**
e *
**
*
e e1 e1
e
**
Species Species richness Composition
*** **
e
e .
*
e e1 e1
e
e
171
Litter
N release (L)
Litter N release (L)
**
Litter Li release (L)
* e
e .
e
e
e
.
e ***
e
e .
e ** .
.
e
***
e
.
*
**
e
e
.
e *
e
e
e
e
e
*
Table 3 Summarized results of ANOVA significance tests for plot and litter species specific tracer transfer among litter species (tracer uptake of unlabelled litter species from labelled litter species in %) and the 15N uptake by phytometer plants (tracer uptake of phytometers from labelled litter in %) at the specific time steps. (U) refers to results of unlabelled leaves. Horizontal lines indicate factors which could be not tested in the respective model or according to data specific circumstances, indicated by 1 (Q. acutissima was excluded from analyses due to very low initial Li contents of labelled litter) and 2 (in addition to 1, no Li uptake found for S. superba leaf litter). Significance levels: P < 0.001 (***), P < 0.01 (**), P < 0.05 (*), P < 0.1 (.). Qu (Q. acutissima), Sa (S. saponaria), Sc (S. superba). Days of decomposition
Factors tested
Litter 15N transfer among species (U)
Litter Li transfer among species (U)
15
plot
Qu
Sa
Sc
plot
12
Species Species richness Composition
***
e
e
e
37
101
171
.
Species Species richness Composition
e
Species Species richness Composition
e
Species Species richness Composition
***
e
e
e
e
model as fixed effects, and block and plot as random effects (Tables 2 and 3). For species specific analyses, the same model was applied without litter species identity. All data were analysed using R 2.13.0 (http://r-project.org) with the packages 'lme4and 'stats. 3. Results 3.1. Effects of litter species identity Litter species identity affected decomposition dynamics across all steps of mass loss, nutrient release (Table 2, Fig. 2, Tables S1eS4), transfer among litter species (Table 3, Fig. 3, Tables S5eS6) and uptake by phytometer plants (Table 3, Fig. 4, Table S7). Litter decomposition rates differed strongly between species (Table 2, P < 0.001 at all individual dates), with highest decomposition rates of S. saponaria leaf litter (k ¼ 8.33 ± 0.47 yr1, corresponding to a mean weight loss after 171 days of 98%), followed by S. superba (3.59 ± 0.28 yr1, 81% mass loss) and Q. acutissima (3.56 ± 0.24 yr1, 81% mass loss). This faster mass loss pattern of S. saponaria was also found for N and 15N loss (Fig. 2), and the 15N uptake by indicator plants (Fig. 4), while Li loss was similar for S. saponaria and S. superba. In contrast, Q. acutissima showed no Li release over time
Qu
Sa
Sc
plot
Qu
Sa
Sc
e
e
e e2 e2
***
e
e
e
e e2 e2
***
e
e
e
e e2 e2
.
e
e
e
e e2 e2
.
e
e
e1
*
.
e
** . ***
e * **
e1
** . ***
e
e
***
e
e
**
*
e1
e
e
N uptake by phytometer plants
e
e
1
. e
which was due to very low initial Li concentrations, leading to the decision to exclude it as a donor species in the statistical analyses of Li transfer among litter species (Table 3, Fig. 2). We observed species specific difference for the 15N and Li uptake of unlabelled litter species from labelled litter which were time-dependent (P < 0.001 at day 12 and 171 of decomposition for 15N and P < 0.01 for day 37 and 101 and P < 0.001 at day 171 for Li). Here, highest nutrient uptake was observed for Q. acutissima, followed by S. saponaria, while no uptake was found for S. superba (Fig. 3). 3.2. Effects of litter species diversity Litter species richness affected mass loss, 15N and Li release (Fig. 5) and Li transfer dynamics (Fig. 6). However, these effects were highly species specific, time-dependent and did not propagate across different processes (Tables 2 and 3). Species richness effects on nutrient transfer among litter species were different from those on mass loss and nutrient release. Litter mass loss in three species mixtures was slower compared to other species richness levels at day 171 of decomposition (P < 0.01; Table 2, Fig. 5). This effect was caused by negative species richness effects on S. saponaria (P < 0.001) and S. superba (P < 0.1) at this sampling
48
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
Fig. 1. Ratio of labelled and unlabelled leaf litter within monocultures, 2- and 3-species mixtures.
Fig. 2. Species specific mass loss and nutrient release of decomposing litter over time, with percent of remaining litter mass, 15N, N and Li after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors.
date. Release of 15N or N was never affected by litter species richness at the plot level. At the species level only 15N release of S. saponaria at day 171 was negatively affected (P < 0.01, Table 2), similar to the pattern found for mass loss. Litter Li release was faster in three species mixtures after 37 (P < 0.01) and 101 days
(P < 0.001) of decomposition due to positive effects of species richness in S. saponaria (P < 0.01) and S. superba (P < 0.1) at day 37 and S. superba (P < 0.05) at day 101 (Table 2). For Li transfer from labelled to unlabelled litter species, more Li was transferred in 2species mixtures at day 37 and 101 of decomposition (P < 0.1),
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
49
Fig. 3. Nutrient transfer among decomposing litter species, shown as percent of 15N and Li uptake of unlabelled litter from the corresponding labelled litter species after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). The nutrient uptake of S. superba is not displayed since no tracer uptake was observed. Displayed are mean values with standard errors. Q. acutissima was excluded as a donor species for Li transfer.
Fig. 4. 15N uptake by phytometer plants from labelled litter species after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors.
partly caused by higher uptake of Q. acutissima at day 37, while no effects were found for 15N transfer (Table 3; Fig. 6). Litter species composition affected mass loss (P < 0.01; day 101) and Li release (P < 0.01; day 37) but strongly depended on decomposition time and litter species identity (Table 2). We observed a consistent higher transfer of Li from labelled S. saponaria litter to unlabelled Q. acutissima leaves across all timesteps, while the 15N transfer to unlabelled litter of Q. acutissima (P < 0.1) and S. superba (P < 0.05) after 12 days was solely slightly affected by composition (Table 3). The 15N uptake by phytometer plants was neither affected by litter species richness nor by composition (Fig. 7, Table 3). 4. Discussion For the first time we traced two essential nutrients for plants though different processes of litter decomposition dynamics along a litter species richness gradient in a field trial. Our results provide new insights into the role of litter species identity and richness on nutrient dynamics in litter mixtures and the subsequent uptake by
plants. 4.1. Effects of litter species identity It is generally accepted that litter decomposition rates are highly species-specific. Litter quality traits (chemical composition and stoichiometry), determine the attractiveness of the food source for decomposer organisms, while leaf shape and toughness affect microhabitat conditions and physical access for decomposers (e.g. €ttenschwiler et al., reviewed in Gartner and Cardon, 2004; Ha 2005a). Species identity effects were also predominant in our study and propagated through all of the investigated parameters characterizing decomposition and nutrient dynamics. For example, litter of S. saponaria showed highest decomposition rates compared to the other species. This pattern was also observed for 15N release and uptake by plants from labelled leaves of this litter species. Leaves of S. saponaria were soft, with the lowest tensile strength of all species (Eichenberg D., personal communication). This speciesspecific leaf toughness has been recognized as an important
50
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
Fig. 5. Mass loss and nutrient release of decomposing litter over time, with percent of remaining litter mass, 15N, N and Li in monocultures, 2- and 3-species mixtures after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors.
Fig. 6. Nutrient transfer among decomposing litter species, shown as percent of 15N and Li transfer of unlabelled litter from the corresponding labelled litter species in 2-, and 3species mixtures after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors. Q. acutissima was excluded as a donor species for Li transfer.
predictor for litter decomposition rates (e.g. Cornelissen et al., 1999; rez-Harguindeguy et al., 2000), as it represents physical barriers Pe (e.g. cuticula, cell walls). Such barriers likely also explain part of species specific effects on nutrient transfer to unlabelled litter. For
example, S. superba showed no signs of nutrient transfer from other species (Fig. 3), which might be due to the leathery leaves with highest tensile strength of all three species (Eichenberg D., personal communication) that could negatively affect colonization by
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
51
Fig. 7. 15N uptake by phytometer plants from labelled monoculture, 2- and 3-species leaf litter after 12, 37, 101 and 171 days of litter decomposition (shifted for better visibility). Displayed are mean values with standard errors.
microorganisms such as fungi or bacteria. This assumption is supported by results on polyphenols of the litter used in our experiment reported by Ristok et al. (2017). Highest polyphenol contents were here found in S. superba, while lowest contents were observes for S. saponaria, which fits well to our data (Table 1). Also nutrient release and transfer dynamics showed different species specific patterns for 15N and Li, with a faster cation release of labelled and higher cation uptake of unlabelled leaves compared to 15N. In contrast to N, Li release was much faster and similar for S. saponaria and S. superba. These results suggest that release dynamics depend on the nutrient type, which is related to the chemical nature of these elements. For example, N is mainly chemically bound in organic molecules, while K and its chemically comparable tracer surrogate Li (Gockele et al., 2014) appear in highly soluble forms, resulting in a fast initial K and slower N release (e.g. Blair, 1988; Briones and Ineson, 1996; Yang et al., 2004; Schreeg et al., 2013). Further, initial nutrient amounts in litter and their stoichiometry affect the attractiveness of this litter species as a food source, which may explain species specific effects. Similar to our results, also Ristok et al. (2017) observed species identity effects on polyphenol composition (only four shared compounds found in S. superba and Q. acutissima) and on phenol and tannin decomposition rates. But also data of green leaves of the same species but another experiment showed species specific contents also for other nutrients (Table S8). These differences likely become less important when the specific nutrient is released very fast. All these results thus support our first hypothesis, in which we expected propagating species identity effects and nutrient specific differences in these highly interconnected steps of litter decomposition processes.
4.2. Effects of litter species richness In contrast to our second hypothesis, we did not observe general positive litter species richness effects on litter decomposition dynamics, and those few effects found were idiosyncratic, species specific and time-dependent. Further, these effects did not propagate through the studied processes of decomposition, which is also contrary to our expectation. However, species diversity mainly influenced Li dynamics, indicating that effects of litter species richness and composition on decomposition processes can be nutrient specific. Litter diversity effects on Li transfer from labelled litter to unlabelled Q. acutissima litter showed no effects on other decomposition processes of this species. These findings suggest
that the amount of nutrients transferred was too low to affect other decomposition processes or that species identity effects may outweigh diversity effects on nutrient transfer. Nutrient transfer among species in decomposing litter mixtures may thus be less important for litter diversity effects on decomposition processes compared to species identity effects. This is also in contrast to our third hypothesis in which we expected that observed richness effects on decomposition dynamics will be also found for the nutrient transfer among litter species. The few observed time and nutrient dependent diversity effects could have been caused by a range of factors. Especially the initial species-specific chemical composition (H€ attenschwiler et al., 2005b) and physical litter traits may affect a range of interconnected mechanisms during decomposition. For example, the physical structure of the litter layer determines the aggregation and hence litter contact areas within mixtures. Large contact areas between leaves likely enhance nutrient release but also transfer rates among different litter species. These contact areas very likely differ with species diversity and change with decomposition time. A dense packing, which usually increases with time, as well as litter traits such as toughness, but also particle size, can further positively affect water-holding capacities (e.g. Naeth et al., 1991; Dirks et al., 2010) and thus litter decomposition (Jin et al., 2013; Makkonen et al., 2013). We thus suggest that an increased water-holding capacity also positively affected cation release. Leaching of soluble cations would then be more sensitive towards litter packing and structural changes in mixtures compared to less soluble dissolved organic N (Hagedorn and Machwitz, 2007). This assumption is supported by our results, which show litter richness effects on Li but not on 15N contents. In contrast to Li, mass loss and 15N release showed time-dependent species diversity effects, with slower mass loss and 15N release in 3-species mixtures at day 171. We do not know if and to which extent this is an artefact, e.g. due to random variation, too low values at the third time step (no increase at the last time step), due to contamination of smaller litter particles with mineral material or by a stronger fungal colonization in diverse mixtures at the last time step. However, it is also possible that such negative effects were caused by inhibiting substances of species which hamper the attractiveness to decomposer of other species in mixtures (Schimel et al., 1998). If one species would release such inhibiting substances, the higher probability to have that species present in mixtures would thus explain an overall negative diversity effect (sampling effect, Tilman et al., 2014). It is likely that these two different
52
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53
mechanisms for highly soluble and moderate soluble nutrients, respectively, were operating in our study system: Water holding capacity might have been important for leaching of the soluble cation, while mass loss and N release were likely hampered by retardants. This is supported by other studies, showing that decomposer organisms are not affected by water holding capacity (e.g. Taylor et al., 2004; Bruder et al., 2011) but by inhibiting substances (Schimel et al., 1998). Although species richness slowed litter and 15N mass loss in three species mixtures after 171 days of decomposition, species richness did not affect the 15N uptake by phytometer plants. This suggests that only strong differences among litter species in mass loss may affect short term uptake rates by plants. Slight diversity effects on mass loss and nutrient release are thus likely buffered by the soil (i.e. nutrients from decomposing litter contribute to an already existent nutrient pool in the soil which likely buffers temporal differences in nutrient input) and need more time and decomposed litter of the respective litter mixtures to develop. Species richness effects on litter decomposition would then not affect nutrient uptake dynamics by plants after a short time span of 171 days. Diversity effects on different ecosystem processes often need time to develop (e.g. Fargione et al., 2007; Reich et al., 2012) while species identity effects occur from the beginning and remain important over time (e.g. Vivanco and Austin, 2008). We observed similar effects in our decomposition experiment for litter mass loss over time. Here, species identity affected the whole decomposition process, while richness effects on mass loss occurred solely at the end of the experiment. Several interconnected processes such as decomposer colonization and interactions among decomposer guilds, and the transfer and release of nutrients or inhibiting substances from in litter mixtures need time to develop and likely increase over time with litter packing. 5. Conclusion We conclude that the use of tracer elements is a powerful technique to investigate the different processes of nutrient dynamics within decomposing litter, ranging from mass loss and associated nutrient release and transfer among litter species, to the subsequent uptake of the mineralized nutrients by plants. We were thus able to demonstrate for the first time the potential impacts of litter species diversity and identity for these different processes involved in decomposition of organic matter. Particularly, our results confirm the dominant effect of species identity and composition - and thus the functional importance of plant traits - not only on mass loss, but also on related nutrient dynamics. In addition, species richness seems to be relevant only at certain points in time for nutrient dynamics in decomposing litter, but not necessarily for the final uptake of N by plants. This latter result supports the prior assumption that litter species mixture effects on decomposition rates may have no direct implications on plant nutrition (e.g. Wardle et al., 1997), at least in subtropical climate zones and for the observed litter species. Plant nutrition is thus rather determined by species specific nutrient contents which are released during decomposition. Nevertheless, our study reinforces the concept of strong biotic control over decomposition as a key process in ecosystem functioning, which opens the possibility to actively use the huge diversity of different tree species in management of ecosystem services provided by forests. Acknowledgements We thank our cooperation partners Zhiqin Pei, Katherina A. Pietsch, David Eichenberg, Junfei Guo, Jessica Gutknecht, Sabine
Both, Manuel Hib, Helge Bruelheide and many other people for their help to establish the experimental setup. We thank Steffen Seitz, Phillip Goebes from the University of Tübingen for soil analysis and Erika Fischer from the Centre for Biological Systems Analysis at the University of Freiburg for IRMs analysis. We are grateful for the help of Ulrike Erhard, Benjamin Plaga, Wolfgang H. Müller and student helpers who made our intensive lab work possible and Angela L. de Avila for comments on a former version of this manuscript. We further thank Barbara Weisbrod and Benjamin Plaga for their help with collecting data presented in Table S8. We are grateful for the funding provided by the German Research Foundation (DFG FOR 891/2). We further thank the Sino-German Centre for Research Promotion for funding a Summer School in Jingdezhen in 2015 (GZ 1146) about scientific writing, where parts of this article were written. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.soilbio.2017.08.007. References Ball, B.A., Hunter, M.D., Kominoski, J.S., Swan, C.M., Bradford, M.A., 2008. Consequences of non-random species loss for decomposition dynamics: experimental evidence for additive and non-additive effects. Journal of Ecology 96, 303e313. http://dx.doi.org/10.1111/j.1365-2745.2007.01346.x. Berg, B., Berg, M.P., Bottner, P., Box, E., Breymeyer, A., Anta, R.C., de, Couteaux, M., €lko € nen, E., Escudero, A., Gallardo, A., Kratz, W., Madeira, M., Ma ~ oz, F., Piussi, P., Remacle, J., Santo, A.V. McClaugherty, C., Meentemeyer, V., Mun de, 1992. Litter mass loss rates in pine forests of Europe and Eastern United States: some relationships with climate and litter quality. Biogeochemistry 20, 127e159. http://dx.doi.org/10.1007/BF00000785. Berg, B., McClaugherty, C., 2008. Plant Litter: Decomposition, Humus Formation, Carbon Sequestration, 2nd. ed. Springer, Berlin. Blair, J.M., 1988. Nutrient release from decomposing foliar litter of three tree species with spicial reference to calcium, magnesium and potassium dynamics. Plant and Soil 110, 49e55. Briones, M.J.I., Ineson, P., 1996. Decomposition of eucalyptus leaves in litter mixtures. Soil Biology and Biochemistry 28, 1381e1388. http://dx.doi.org/10.1016/ S0038-0717(96)00158-7. Bruder, A., Chauvet, E., Gessner, M.O., 2011. Litter diversity, fungal decomposers and litter decomposition under simulated stream intermittency: litter diversity, stream intermittency and decomposition. Functional Ecology 25, 1269e1277. http://dx.doi.org/10.1111/j.1365-2435.2011.01903.x. Caner, L., Zeller, B., Dambrine, E., Ponge, J.-F., Chauvat, M., Llanque, C., 2004. Origin of the nitrogen assimilated by soil fauna living in decomposing beech litter. Soil Biology and Biochemistry 36, 1861e1872. Cardinale, B.J., Duffy, J.E., Gonzalez, A., Hooper, D.U., Perrings, C., Venail, P., Narwani, A., Mace, G.M., Tilman, D., Wardle, D.A., Kinzig, A.P., Daily, G.C., Loreau, M., Grace, J.B., Larigauderie, A., Srivastava, D.S., Naeem, S., 2012. Biodiversity loss and its impact on humanity. Nature 486, 59e67. http:// dx.doi.org/10.1038/nature11148. Cardisch, G., Giller, K.E., 1996. Driven by Nature: Plant Litter Quality and Decomposition, first ed. CABI, Wallingford, Oxon, UK. Chapman, K., Whittaker, J.B., Heal, O.W., 1988. Proceedings of a workshop on interactions between soil-inhabiting invertebrates and microorganisms in relation to plant growth metabolic and faunal activity in litters of tree mixtures compared with pure stands. Agriculture, Ecosystems & Environment 24, 33e40. http://dx.doi.org/10.1016/0167-8809(88)90054-0. rez-Harguindeguy, N., Díaz, S., Grime, J.P., Marzano, B., Cornelissen, J.H., Pe Cabido, M., Vendramini, F., Cerabolini, B., 1999. Leaf structure and defence control litter decomposition rate across species and life forms in regional floras on two continents. New Phytologist 143, 191e200. Dierschke, H., 1994. Pflanzensoziologie: Grundlagen und Methoden, first ed. UTB, Stuttgart, Stuttgart. Dirks, I., Navon, Y., Kanas, D., Dumbur, R., Grünzweig, J.M., 2010. Atmospheric water vapor as driver of litter decomposition in Mediterranean shrubland and grassland during rainless seasons. Global Change Biology 16, 2799e2812. http://dx.doi.org/10.1111/j.1365-2486.2010.02172.x. Fahey T.J., Yavitt, J.B., Sherman, R.E., Groffman, P.M., Fisk, M.C., Maerz, J.C., 2011. Transport of carbon and nitrogen between litter and soil organic matter in a Northern hardwood forest. Ecosystems 14, 326e340. http://dx.doi.org/10.1007/ s10021-011-9414-1. Fargione, J., Tilman, D., Dybzinski, R., Lambers, J.H.R., Clark, C., Harpole, W.S., Knops, J.M.H., Reich, P.B., Loreau, M., 2007. From selection to complementarity: shifts in the causes of biodiversityeproductivity relationships in a long-term biodiversity experiment. Proceedings of the Royal Society of London B:
K.N. Leppert et al. / Soil Biology & Biochemistry 115 (2017) 44e53 Biological Sciences 274, 871e876. http://dx.doi.org/10.1098/rspb.2006.0351. Fitter, A.H., 1986. Spatial and temporal patterns of root activity in a species-rich alluvial grassland. Oecologia 69, 594e599. http://dx.doi.org/10.1007/ BF00410368. Frey, S.D., Six, J., Elliott, E.T., 2003. Reciprocal transfer of carbon and nitrogen by decomposer fungi at the soilelitter interface. Soil Biology and Biochemistry 35, 1001e1004. http://dx.doi.org/10.1016/S0038-0717(03)00155-X. Fyles, J.W., Fyles, I.H., 1993. Interaction of Douglas-fir with red alder and salal foliage litter during decomposition. Canadian Journal of Forest Research 23, 358e361. http://dx.doi.org/10.1139/x93-052. Gartner, T.B., Cardon, Z.G., 2004. Decomposition dynamics in mixed-species leaf litter. Oikos 104, 230e246. http://dx.doi.org/10.1111/j.0030-1299.2004.12738.x. Gockele, A., Weigelt, A., Gessler, A., Scherer-Lorenzen, M., 2014. Quantifying resource use complementarity in grassland species: a comparison of different nutrient tracers. Pedobiologia 57, 251e256. http://dx.doi.org/10.1016/ j.pedobi.2014.09.001. Hagedorn, F., Machwitz, M., 2007. Controls on dissolved organic matter leaching from forest litter grown under elevated atmospheric. Soil Biology and Biochemistry 39, 1759e1769. http://dx.doi.org/10.1016/j.soilbio.2007.01.038. €ttenschwiler, S., Tiunov, A.V., Scheu, S., 2005a. Biodiversity and litter decompoHa sition in terrestrial ecosystems. Annual Review of Ecology, Evolution, and Systematics 36, 191e218. http://dx.doi.org/10.1146/ annurev.ecolsys.36.112904.151932. €ttenschwiler, S., Tiunov, A.V., Scheu, S., 2005b. Biodiversity and litter decompoHa sition in terrestrial ecosystems. Annual Review of Ecology, Evolution, and Systematics 36, 191e218. http://dx.doi.org/10.1146/ annurev.ecolsys.36.112904.151932. Hooper, D.U., Chapin, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H., €, H., Symstad, A.J., Lodge, D.M., Loreau, M., Naeem, S., Schmid, B., Set€ ala Vandermeer, J., Wardle, D.A., 2005. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecological Monographs 75, 3e35. http://dx.doi.org/10.1890/04-0922. IUSS Working Group WRB, 2015. World Reference Base for Soil Resources 2014, Update 2015. International soil classification system for naming soils and creating legends for soil maps. Jin, V.L., Haney, R.L., Fay, P.A., Polley, H.W., 2013. Soil type and moisture regime control microbial C and N mineralization in grassland soils more than atmospheric CO 2-induced changes in litter quality. Soil Biology and Biochemistry 58, 172e180. Johnson, D.W., Todd, D.E., 1998. Effects of Harvesting Intensity on Forest Productivity and Soil Carbon Storage in a Mixed Oak Forest. Kaneko, N., Salamanca, E., 1999. Mixed leaf litter effects on decomposition rates and soil microarthropod communities in an oakepine stand in Japan. Ecological Research 14, 131e138. http://dx.doi.org/10.1046/j.1440-1703.1999.00292.x. Lummer, D., Scheu, S., Butenschoen, O., 2012. Connecting litter quality, microbial community and nitrogen transfer mechanisms in decomposing litter mixtures. Oikos 121, 1649e1655. http://dx.doi.org/10.1111/j.1600-0706.2011.20073.x. Makkonen, M., Berg, M.P., van Logtestijn, R.S.P., van Hal, J.R., Aerts, R., 2013. Do physical plant litter traits explain non-additivity in litter mixtures? A test of the improved microenvironmental conditions theory. Oikos 122, 987e997. http:// dx.doi.org/10.1111/j.1600-0706.2012.20750.x. McArthur, J. m, 1994. Recent trends in strontium isotope stratigraphy. Terra Nova 6, 331e358. http://dx.doi.org/10.1111/j.1365-3121.1994.tb00507.x. McTiernan, K.B., Ineson, P., Coward, P.A., 1997. Respiration and nutrient release from tree leaf litter mixtures. Oikos 78, 527. http://dx.doi.org/10.2307/3545614. Naeth, M.A., Bailey, A.W., Chanasyk, D.S., Pluth, D.J., 1991. Water holding capacity of litter and soil organic matter in mixed prairie and fescue grassland ecosystems of Alberta. Journal of Range Management 44, 13e17. http://dx.doi.org/10.2307/ 4002630. rez-Harguindeguy, N., Díaz, S., Cornelissen, J.H., Vendramini, F., Cabido, M., Pe
53
Castellanos, A., 2000. Chemistry and toughness predict leaf litter decomposition rates over a wide spectrum of functional types and taxa in central Argentina. Plant and Soil 218, 21e30. http://dx.doi.org/10.1023/A:1014981715532. Peterson, D.L., Rolfe, G.L., 1982. Nutrient dynamics and decomposition of litterfall in floodplain and upland forests of central illinois. Forest Science 28, 667e681. Reich, P.B., Tilman, D., Isbell, F., Mueller, K., Hobbie, S.E., Flynn, D.F.B., Eisenhauer, N., 2012. Impacts of biodiversity loss escalate through time as redundancy fades. Science 336, 589e592. http://dx.doi.org/10.1126/science.1217909. Ristok, C., Leppert, K.N., Franke, K., Scherer-Lorenzen, M., Niklaus, P.A., Wessjohann, L.A., Bruelheide, H., 2017. Leaf litter diversity positively affects the decomposition of plant polyphenols. Plant and Soil 416, 1e13. http://dx.doi.org/ 10.1007/s11104-017-3340-8. 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. Biogeochemistry 42, 221e234. http://dx.doi.org/10.1023/A: 1005911118982. €ttenschwiler, S., 2007. Nitrogen transfer between decomposing Schimel, J.P., Ha leaves of different N status. Soil Biology and Biochemistry 39, 1428e1436. http://dx.doi.org/10.1016/j.soilbio.2006.12.037. Schreeg, L.A., Mack, M.C., Turner, B.L., 2013. Nutrient-specific solubility patterns of leaf litter across 41 lowland tropical woody species. Ecology 94, 94e105. Seastedt, T.R., 1984. The role of microarthropods in decomposition and mineralization processes. Annual Review of Entomology 29, 25e46. http://dx.doi.org/ 10.1146/annurev.en.29.010184.000325. Seitz, S., Goebes, P., Zumstein, P., Assmann, T., Kühn, P., Niklaus, P.A., Schuldt, A., Scholten, T., 2015. The influence of leaf litter diversity and soil fauna on initial soil erosion in subtropical forests. Earth Surface Processes and Landforms 40, 1439e1447. http://dx.doi.org/10.1002/esp.3726. € ter, D., Pflug, A., Wolters, V., 2004. Response of different decomTaylor, A.R., Schro poser communities to the manipulation of moisture availability: potential effects of changing precipitation patterns. Global Change Biology 10, 1313e1324. http://dx.doi.org/10.1111/j.1365-2486.2004.00801.x. Tilman, D., Isbell, F., Cowles, J.M., 2014. Biodiversity and ecosystem functioning. Annual Review of Ecology, Evolution, and Systematics 45, 471e493. http:// dx.doi.org/10.1146/annurev-ecolsys-120213-091917. Tiunov, A.V., 2009. Particle size alters litter diversity effects on decomposition. Soil Biology and Biochemistry 41, 176e178. http://dx.doi.org/10.1016/ j.soilbio.2008.09.017. Vivanco, L., Austin, A.T., 2008. Tree species identity alters forest litter decomposition through long-term plant and soil interactions in Patagonia, Argentina. Journal of Ecology 96, 727e736. http://dx.doi.org/10.1111/j.1365-2745.2008.01393.x. Wardle, D.A., Bonner, K.I., Nicholson, K.S., 1997. Biodiversity and plant litter: experimental evidence which does not support the view that enhanced species richness improves ecosystem function. Oikos 79, 247. http://dx.doi.org/10.2307/ 3546010. €ber, W., Ma, K., Nadrowski, K., Yang, X., Bauhus, J., Both, S., Fang, T., H€ ardtle, W., Kro Pei, K., Scherer-Lorenzen, M., Scholten, T., Seidler, G., Schmid, B., von Oheimb, G., Bruelheide, H., 2013. Establishment success in a forest biodiversity and ecosystem functioning experiment in subtropical China (BEF-China). European Journal of Forest Research 132, 593e606. http://dx.doi.org/10.1007/s10342-0130696-z. Yang, Y.S., Guo, J.F., Chen, G.S., Xie, J.S., Cai, L.P., Lin, P., 2004. Litterfall, nutrient return, and leaf-litter decomposition in four plantations compared with a natural forest in subtropical China. Annals of Forest Science 61, 465e476. Zeller, B., Colin-Belgrand, M., Dambrine, E., Martin, F., 2001. Fate of nitrogen released from 15N-labeled litter in European beech forests. Tree Physiology 21, 153e162. Zeller, B., Colin-Belgrand, M., Dambrine, E., Martin, F., Bottner, P., 2000. Decomposition of 15N-labelled beech litter and fate of nitrogen derived from litter in a beech forest. Oecologia 123, 550e559. http://dx.doi.org/10.1007/PL00008860.