• biodiversity;
  • functional traits;
  • legumes;
  • litter quality


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    A number of biodiversity experiments have shown that plant diversity plays a significant role for ecosystem functioning. However, diversity effects on processes involving multi-trophic interactions such as litter decomposition are rather rare. In these experiments, plant diversity is usually categorized into taxonomic units or functional groups. Continuous measures of functional diversity that are based on measurable traits, in contrast, may be a more flexible way to analyse the functional significance of biodiversity.
  • 2
    Litter decomposition is a key process in ecosystem biogeochemistry. To understand the consequences of altered biodiversity for ecosystem functioning, it is thus crucial to quantify any potential diversity effects on decomposition processes.
  • 3
    I performed several complementary decomposition trials within the BIODiversity and Ecological Processes in Terrestrial Herbaceous ecosystems (BIODEPTH) experiment, which established a gradient of plant species richness and number of functional groups. I hypothesized that decomposition rates increase with increasing plant diversity due to non-additive litter mixing effects and more favourable microenvironmental conditions.
  • 4
    Decomposition rates of both standard materials and community-specific litter increased with the number of functional groups and with a continuous measure of functional diversity. Species richness, in contrast, had no or rather small positive effects on decomposition. Presence of nitrogen-fixing legumes strongly enhanced decomposition, via effects on both litter quality and on the decomposition microenvironment.
  • 5
    The predominance of functional diversity rather than of species richness effects suggests that profound knowledge on functional attributes of plant species is the key to understand and to predict biodiversity–decomposition relationships.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

During the last 15 years, a major research effort in ecology has been related to the question whether biodiversity matters for ecosystem functioning, that is, for the processes and properties of ecosystems. There is now ample evidence for such biodiversity–ecosystem functioning relationships, with generally strong positive effects of plant diversity on primary production, accompanied by higher nutrient uptake with depletion of soil resources in diverse systems (see recent reviews by Hooper et al. 2005; Balvanera et al. 2006). In contrast, significant effects on other ecological processes – especially those involving multi-trophic interactions such as litter decomposition – are rare or lag behind above-ground responses.

Decomposition of organic matter is a key process in ecosystem's biogeochemistry, replenishing the pool of plant-available soil nutrients and releasing photosynthetically fixed carbon back to the atmosphere. It is thus crucial to estimate diversity effects on litter decomposition to understand the consequences of changes in plant biodiversity for ecosystem functioning. A first step in that direction is to compare decomposition rates of litter from different species under identical environmental conditions. Such studies showed that species do differ widely in their decomposition rates (Cornelissen 1996). In a further step, decomposition in pure and mixed litters can be compared to elucidate any kind of mixing effects. Apparently, non-additive litter mixing effects do prevail, that is, litter mass loss in mixtures is higher or lower than in pure litters (Gartner & Cardon 2004), suggesting that some interactions among different litter species do affect litter decomposition (Hättenschwiler, Tiunov & Scheu 2005). The third step tests whether there is an effect of litter diversity beyond the pure vs. mixture contrast by analysing decomposition along gradients of litter diversity. In these experiments no general relationship between diversity and decomposition processes has been found so far (Mikola, Bardgett & Hedlund 2002; Wardle & Van Der Putten 2002; Hättenschwiler et al. 2005).

One of the reasons for the lack of a predictable diversity effect on decomposition is the multitude of pathways through which changes in species diversity could affect decomposition. Litter decomposition is influenced by microclimatic conditions, by the diversity, composition and activity of the decomposer community (soil fauna, fungi and bacteria) and by chemical and physical characteristics of litter (Swift, Heal & Anderson 1979; Cadisch & Giller 1997). It can be assumed that all these factors are influenced directly or indirectly by the composition and structure of the plant community (Wardle 2002; Hättenschwiler et al. 2005; Prescott 2005); for example, through altered quantity and quality of litter inputs influencing decomposer biomass, composition and activity, or through differences in canopy architecture affecting evapotranspiration and hence microclimate. Additionally, the spatial and temporal pattern of litter production could well be affected by plant diversity.

In biodiversity experiments, ecosystem responses are usually related to different pre-defined levels of taxonomic diversity (species richness) or number of functional groups that are classified a priori based on differences in life-history strategies or certain morphological and physiological traits (Schmid et al. 2002). More recently, continuous measures of functional diversity1 have increasingly been used in analysing biodiversity–ecosystem functioning relationships because categorical classifications have several limitations, for example, information about variation of functional trait diversity within groups is lost (Ricotta 2005; Petchey & Gaston 2006). Those diversity measures offer the possibility to select certain traits that are of particular relevance for the process under study. Thus, indices of functional diversity based on selected, measurable traits provide a quantitative and flexible way to test more specific hypotheses about the functional significance of plant diversity.

The work presented here is an attempt to disentangle some of the above-mentioned pathways of diversity effects on decomposition of organic matter, applying both categorical and continuous measures of plant diversity for analyses. I hypothesized that decomposition rates increase with increasing plant diversity due to non-additive litter mixing effects and more favourable microenvironmental conditions. Further, I expected higher explanatory power of measures of functional diversity than of measures of taxonomic diversity because plant functional attributes strongly control ecological processes. In consequence, community composition should also explain a large portion of the variability in decomposition rates. The study was part of the pan-European project BIODiversity and Ecological Processes in Terrestrial Herbaceous ecosystems (BIODEPTH), where communities differing in plant diversity were established from seed (Hector et al. 1999; Spehn et al. 2005).

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Site properties and the experimental design are described in detail in Scherer-Lorenzen et al. (2003), so I only report the most important features here.


The work was done during 1996–1998 at the German location of the BIODEPTH network near Bayreuth (49°55′ N, 11°35′ E, altitude 355 m a.s.l.). The soil is a loamy to sandy stagnic gleysol with a pH (CaCl2) = 5·65 ± 0·20, and low carbon and nitrogen contents (Ctotal = 0·78 ± 0·06%; Ntotal = 0·08 ± 0·01%). During the 3-year study period, annual precipitation varied considerably (582–816 mm), with a remarkable dry summer in 1998 and an exceptional wet October in the same year. Mean annual temperatures ranged from 6·8 °C (1996) to 8·2 °C (1998).

experimental design

The experiment was established on former arable land. In early spring 1996, the soil was harrowed and the seedbank was eliminated by in situ steam sterilization. Composition of experimental communities was directly manipulated to vary species richness and functional group diversity (a priori classification: grasses, legumes, non-leguminous herbs; Table 1). From a pool of 31 grassland species (Arrhenatheretum association), communities were assembled from seed by independent random draws, with the constraint that all multi-species communities had to contain grasses. Each diversity level was replicated with several different mixtures. All mixtures were allocated in a random block design, yielding a total of 60 experimental plots of 2 × 2 m in size. Total seed density was 2000 viable seeds per m2, divided equally among all species following a substitutive replacement series design. Unsown seedlings were continuously weeded and biomass was cut at 5 cm and removed twice a year (June, September. See Hector et al. 1999 and Spehn et al. 2005 for results on biomass production). The plots were not fertilized during the whole experimental period.

Table 1.  Experimental design and replication of richness levels. Table entries are the number of different plant mixtures per diversity level, arranged according to functional groups
Functional group richnessSpecies richnessSubtotal
 1 2 4 816
1Grasses (G)5211 − 9
1Legumes (L)2 − 2
1Herbs (H)3 − 3
2G + L211 − 4
2G + H311 − 5
3G + L + H22 3 7
Subtotal (replicates per block)10755 330
Grand total (two blocks)20141010 660

decomposition trials

To elucidate different ways of plant diversity effects on litter decomposition, I conducted three complementary experiments: First, I used standard materials to test for the effects of the microenvironment provided by each experimental community (diversity–microenvironment experiment). Second, I analysed effects of varying litter composition and quality by inserting litterbags of each community into a homogeneous grassland patch outside the experimental field (litter composition experiment). Finally, with community-specific litterbags that were buried in their corresponding communities, I studied the combined effects of the microenvironment and litter quality (combined microenvironment and composition/quality experiment).

diversity–microenvironment experiment

I measured dry weight loss of cotton strips as an indicator of short-term decomposition of organic matter. Strips of 5 × 12 cm of a special cotton fabric (Shirley Soil Burial Test Fabric, c. 95% cellulose; initial nitrogen concentration of 0·09%) were sealed in a nylon net (mesh size: 1·5 mm) to protect them from physical damage and to facilitate retrieval. Three strips per plot were inserted vertically to a depth of 10 cm using a spade. The soil was then pressed back against the strip to insure full contact between fabric and soil. The cotton strip assay was done twice, both in the second (1997) and third year (1998) of the experiment. The strips remained c. 10 weeks in the field during autumn, that is, from 29 August 1997 until 13 November 1997 and from 1 September 1998 until 17 November 1998, respectively. After retrieval, strips were washed carefully, dried at 80 °C, and weighed. Decomposition rate was calculated as loss of dry weight [g g−1 d−1].

To measure long-term decomposition in the upper soil layer, I determined weight loss of wooden birch sticks. After drying at 80 °C and weighing, one target stick (12 × 9 × 2 cm, ‘ice-cream sticks’) was bound between two identical sticks to facilitate retrieval. The bundles were buried vertically to a depth of 10 cm. Four stick bundles were used per plot and buried for more than 2 years (22 November 1996 until 15 December 1998). After retrieval, the target sticks were washed carefully, dried and weighed and mean decomposition was obtained from differences in dry weight [g g−1 d−1].

litter composition experiment

For the litterbag experiments, I collected fully senescent leaves and stems of all species present in each community during the biomass harvests. The samples were mixed thoroughly and a subsample was ground for total carbon and nitrogen analysis (C/N Analysator, CARLO ERBA NA 1500, Mailand, Italy). A second subsample was oven-dried (80 °C) and an aliquot of 1 g was sealed in bags (5 × 5 cm) made of 0·5 mm nylon mesh. The proportion of litter species within each bag represented the abundance of species within each mixture at the time of biomass harvest in June 1998. The bags were moistened and placed vertically into the soil such that the upper part of the bag protruded above the soil surface. Four litterbags per experimental community were inserted randomly within a homogeneous 2 × 2 m patch of an adjacent Arrhenatherum meadow, providing a uniform environment. The litterbags remained in the soil during the same time period as the cotton strips in 1998. After retrieval, the remaining litter was washed carefully, dried at 80 °C for 24 h and weighed, and decomposition rates were obtained by dry weight loss [g g−1 d−1].

combined microenvironment and composition experiment

In this experiment, I applied the same litterbag methodology as described for the litter composition experiment. Bags with community-specific litter from the October harvest 1997 were placed in their corresponding plots, and remained in situ during the same time period as the cotton strips in 1997.


Data were analysed using a general-linear modelling approach to test for the effects of blocks (B), species richness (S), number of functional groups (F) and all other residual differences between mixtures (M). The model terms were fitted sequentially (Type I SS; spss 15·0), and the effects of B and M were considered random factors. Effects of S and F were tested against the individual mixture effects. The variance among mixtures was then compared with the residual variance among plots. Thus, the problem of confounding effects of species identity and species number or number of functional groups was avoided (Schmid et al. 2002). Interactions between S and F were never significant, so they were excluded from the final model. To test linearity regression function of S and F, the two factors were split into a test for linearity (for the logarithm of species number or for the functional group number) and a test for the deviation from linearity. Due to the non-orthogonality with respect to S and F the variance ratios were dependent on the order in which they were added to the statistical model. Consequently, all analyses were done twice, with both orders of fitting: one of which maximizes the effects of S, and the other of F. Since the presence of particular functional groups such as nitrogen-fixing legumes or grasses had profound effects on ecosystem processes associated with nutrient cycling in this experiment, the effects of each group were analysed as separate contrasts within mixtures. However, the design was unbalanced with respect to the presence and absence of particular functional groups (Table 1), so these analyses have to be interpreted cautiously.

In order to test effects of functional trait diversity on decomposition processes, I calculated quadratic diversity Q as an index of functional diversity. Q is an extension of the Gini–Simpson index of diversity and incorporates functional dissimilarities between species as well as relative abundances into one measure of functional diversity (Ricotta 2005). Q is calculated as

  • image

where S is the number of species, piwith i = 1, ... , S are the relative abundances and dij describe the distance between species i and j in the community, that is, their dissimilarity. A matrix of functional trait values of all species present in each community was used to determine the functional dissimilarities between species in a multidimensional trait space. I used six functional effect traits (sensu Lavorel & Garnier 2002) that potentially influence decomposition processes both via effects on the microenvironment (growth form, leaf size, leaf inclination; all affecting light transmittance, shading and hence soil moisture) and via effects on litter quality (seasonality of foliage, C : N ratio, specific leaf area; see Cornelissen & Thompson 1997). The minimal value of Q is zero and occurs if all species are identical with respect to the functional traits measured (i.e. dij = 0 for all i,j), or if monocultures are considered. The relationship between functional diversity and decomposition processes was analysed by linear regression.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

standardized materials: the diversity-microenvironment experiment

In both years, an increase of functional group richness resulted in a significant linear increase in decomposition of cotton strips, and a marginally significant increase in wood decomposition (Fig. 1a, Table 2, wood: P = 0·093 if fitted after species richness). If legumes were present, standard materials decomposed significantly faster (Fig. 1b,c,d andTable 2) and within 10 weeks of exposure, up to 90% of the cotton material was lost. In contrast, presence of grasses led to a decrease in decomposition rates. In mixtures composed of two or more species, decomposition rates increased linearly with functional diversity (Fig. 2a,b,c), although Q only explained between 10% and 15% of the total variation.


Figure 1. (a) Effects of functional group richness on decomposition rate of cotton strips and wooden sticks as test materials for short- and long-term soil decomposing activity. Note that decomposition rates of wood (grey symbols) are one order of magnitude lower than that of cotton decomposition (black symbols). Symbols are slightly staggered to improve readability. Effects of the presence of legumes, grasses and herbs on decomposition of cotton strips in 1997 (b) and 1998 (c), and of wooden sticks (d) averaged across all diversity levels; axis scales are same as in (a). Means ± SE.

Download figure to PowerPoint

Table 2. anova summary of the diversity–microenvironment experiment with standard materials testing the effects of diversity on short- and long-term decomposition. Functional group richness was fitted prior to species richness. Bold numbers represent significance with P ≤ 0·05; italic numbers represent marginal significance with P ≤ 0·1
 Cotton (short-term)Wood (long-term)
Source of variationdfF-valueP% SSF-valueP% SSF-valueP% SS
Block1 1·4530·238 1·21·2640·270 1·6 3·8590·061 7·7
Functional group richness (linear)1 5·1680·03311·95·8460·02411·9 1·2600·273 1·9
Functional group richness (deviation)1 0·0080·999 0·00·0080·999 0·0 0·0160·997 0·0
Species richness (log-linear)1 1·2870·268 3·00·0600·808 0·1 0·9280·345 1·4
Species richness (deviation)3 1·0820·309 7·50·5350·472 3·3 0·8930·355 4·1
Mixture23 2·6890·00753·11·6110·11246·6 0·7730·73135·3
Legume contrast112·1220·00218·96·3270·02010·411·7730·00212·3
Grass contrast1 4·1600·054 8·44·1970·053 7·5 1·0690·312 1·6
Herb contrast1 1·5590·225 3·50·4360·516 0·9 3·7390·066 5·1

Figure 2. Decomposition rates of cotton strips in 1997 (a) and 1998 (b), and of wood (c) as a function of functional diversity Q in communities with more than one species. Symbols correspond to those of Fig. 1.

Download figure to PowerPoint

In contrast, cotton and wood decomposition was not significantly influenced by species richness (Table 2), even if fitted before the effect of functional groups, although a trend of constant increase from the monoculture to the eight-species level could be observed for cotton (data not shown). Generally, very high within-treatment variability could be observed, that is, differences among mixtures other than related to species richness and functional group richness had highly significant effects on decomposition (significant mixture term in Table 2).

litter decomposition: the composition and the combined microenvironment–composition experiment

Generally, litter decomposition increased linearly with increasing functional group richness (Fig. 3a, Table 3) and with increasing functional diversity Q (Fig. 4a,b). In both experiments, presence of litter from legumes led to higher decomposition rates in comparison to litter without legumes (Fig. 3b,c and Table 3). Grasses, in contrast, reduced decomposition rates, especially in the litter composition experiment (homogeneous environment). In the litter composition experiment, decomposition increased with functional diversity even in plots without legumes (P = 0·005, r2 = 0·53).


Figure 3. (a) Effects of functional group richness on decomposition rate of community-specific litterbags. Litterbags were placed either in a homogeneous environment (litter composition experiment) or in their plot-specific environment (combined microenvironment and litter composition experiment). Symbols are slightly staggered to improve readability. Effects of the presence of legumes, grasses and herbs on decomposition of litterbags placed in homogeneous (b) or plot-specific environment (c) averaged across all diversity levels; axis scales are same as in (a). Means ± SE.

Download figure to PowerPoint

Table 3. anova summary of the litter composition and the combined microenvironment and composition experiments with litterbags. Functional group richness was fitted prior to species richness. *df for mixture term is 19 for the litter composition and 21 for the combined experiment. Bold numbers represent significance with P ≤ 0·05; italic numbers represent marginal significance with P ≤ 0·1
Source of variationdfLitter compositionMicroenvironment and composition
F-valueP% SSF-valueP% SS
Block10·8020·379 0·7
Functional group richness (linear)1 4·7180·04317·25·2010·03315·1
Functional group richness (deviation)1 0·4100·748 1·50·4970·688 1·4
Species richness (log-linear)1 0·5780·456 2·10·0150·904 0·0
Species richness (deviation)3 0·2320·636 2·50·1320·720 1·2
Mixture19/21*11·003< 0·00169·33·3930·00261·0
Legume contrast1 6·6450·01918·73·0400·097 8·1
Grass contrast123·356< 0·00139·23·6300·071 9·4
Herb contrast1 0·0060·938 0·00·0280·869 0·1

Figure 4. Litter decomposition rates as a function of functional diversity Q in communities with more than one species. Litterbags were placed either in a homogeneous environment (a; litter composition experiment) or in their plot-specific environment (b; combined microenvironment and litter composition experiment). Symbols correspond to those of Fig. 3.

Download figure to PowerPoint

Plant species richness had no significant effect on litter decomposition when litter was placed within the homogeneous environment (Table 3). However, decomposition increased slightly with increasing number of species in the plot-specific environment; a trend that was marginally significant (P = 0·076 if fitted prior to functional groups). Parallel to the findings of the experiments with standard materials, within-treatment variability was very high and consequently, residual differences among mixtures were large (Table 3).

Legumes also influenced decomposition via litter quality. Litter of legume plots had significantly lower C : N ratios than litter of plots without them (P1997/1998 < 0·001, two-tailed t-test). Decomposition rates were negatively related to the C : N ratio of litter: the lower the C : N ratio, the faster the litter decomposed. This relationship was slightly weaker in the combined microenvironment–composition experiment (P = 0·006, r2 = 0·15) than if litter was placed in a homogeneous environment (P < 0·001, r2 = 0·31), indicating that community composition of the experimental plots influenced decomposition rates through several mechanisms.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

plant diversity effects on decomposition

I observed a general increase of decomposition of both standard materials and plant litter with increasing functional diversity, both if categorizized to a priori number of functional groups and if used as a continuous measure of trait dissimilarities (functional diversity Q). In contrast, effects of species numbers on decomposition of both standard organic materials and leaf litter were absent or rather small, with a slight increase in decomposition rates with more species being present. This finding is consistent with the majority of other experimental studies in grasslands that showed no or idiosyncratic effects of plant species richness on decomposition (e.g. Wardle, Bonner & Nicholson 1997; Spehn et al. 2000; Knops, Wedin & Tilman 2001). Positive species diversity effects have only been reported by Hector et al. (2000), Bardgett & Shine (1999) and at the Greek site of the BIODEPTH project (Spehn et al. 2005).

The mechanistic explanation for these stronger effects of functional diversity than of species richness is that functional attributes of organisms affect ecological processes, rather than taxonomic identity (Díaz & Cabido 2001; Hooper et al. 2002). Hence, both the functional dissimilarity among species in a community, and the occurrence of particular traits with strong effects on the process under study are relevant for ecosystem functioning: First, as indices of functional diversity such as Q measure the differences in functional traits between species within a community, increasing functional diversity corresponds to an increase in the ‘functional trait space’ covered by a community (Ricotta 2005; Petchey & Gaston 2006). The positive effect of Q on decomposition processes reported here, therefore, implies that such differences between species matters for ecosystem functioning. Second, species composition – that is, the combination of certain functional traits – and the occurrence of particular traits with large effects on the processes under study usually explain a large variation of the biodiversity–ecosystem functioning relationship. In this study, a strong effect of particular traits was mainly caused by the presence of legumes, which enhanced decomposition rates (see ‘Effects of litter composition and quality’). It has to be noted, though, that the proportion of communities containing legumes unavoidably increased with the number of functional groups (see Table 1), which essentially represents a sampling effect (Huston 1997) that may partly account for the results. However, in the case of the litter composition experiment, decomposition rates were also positively related to functional diversity Q if only plots without legumes were analysed separately, suggesting that biodiversity can affect decomposition beyond the legume sampling effect. Sampling effects may also occur if the presence of other species with a low C : N ratio, or any other trait directly affecting decomposition rates, would result in both an increase of Q and of decomposition rates, thus confounding Q with the inclusion of species that promote decomposition. However, according to Petchey & Gaston (2006), only traits important for the function of interest should be included in the calculation of functional diversity indices, so that sampling effects due to direct action of functional effect traits may indeed become critical.

effects of litter composition and quality

The strong effects of the specific community composition found here are consistent with other studies that report large species-specific effects on litter decomposition (e.g. Wardle et al. 1997). Depending on the functional attributes of species added or lost from a community, both positive and negative changes in decomposition rates may occur and average rates could be independent of plant diversity. In this experiment, the C : N ratio of the litter mixture explained a large proportion of the observed variation in decomposition rates, which predominately was driven by the presence of nitrogen-fixing legumes with low C : N ratios. It is known from many studies that litter species with high nitrogen concentrations and low C : N ratios show high early decomposition rates and may stimulate decay of more recalcitrant litters (Swift et al. 1979; Cadisch & Giller 1997; Hättenschwiler et al. 2005; Prescott 2005). The occurrence of high-nitrogen legume plant residues or exudates might have increased soil fauna abundance or activity in general and facilitated decay of more recalcitrant compounds, such as grass litter. In this experiment, density and diversity of earthworms was higher in communities containing legumes (Gastine, Scherer-Lorenzen & Leadley 2003), although their contribution to the observed results might have been minimal due to the mesh size used. The low-nitrogen standard materials cotton and wood also decomposed faster in communities with legumes, supporting the nutrient transfer hypothesis. Seastadt (1984) suggested that nutrient release from rapidly decaying species may result in a fertilizer effect and stimulate decomposition of adjacent, recalcitrant species. However, direct evidence for such nutrient transfer among litter types is rather weak (Hättenschwiler et al. 2005; Scherer-Lorenzen, Bonilla & Potvin 2007) and only isotope studies, for example, with 15N labelled litters, would give direct support for or against this hypothesis.

Generally, litter quality was related to the classification of functional groups with a decreasing decay rate in the order legumes (low C : N ratio), non-nitrogen-fixing herbs (fragile litter) and grasses (sclerenchymatic litter). Beside indicators for substrate quality, such as C : N ratio, content of lignin, phosphorus, or lignin : N ratio, physical characteristics of litter are also influencing decomposition. For example, litter mixtures that had high C : N ratios but also high decomposition rates in the homogeneous environment included the herbs Geranium pratense L. or Ranunculus acris L. These species produce very fragile litter in contrast to the more sclerenchymatic one of most grasses. Alopecurus pratensis L., for example, had the lowest decomposition rates, although its C : N ratio did not differ from that of other grasses, showing that other chemical compounds – such as polyphenols or silicium, which act as defence against fungi and herbivores – together with other physical characteristics of the litter – such as surface properties, toughness and particle size – are additional important factors determining litter quality and hence decomposition rates (Cornelissen & Thompson 1997; Hättenschwiler & Vitousek 2000). Given the big differences in morphological and chemical attributes of different litter species, it is thus not surprising that community composition usually is the major determinant of decay rates.

effects of microenvironmental conditions

Beside the more direct effects through litter composition and quality, changes in diversity may indirectly influence decomposition through effects on abiotic soil conditions. Both above-ground biomass and leaf area index (LAI) of the experimental communities strongly increased with increasing plant diversity in 1997 and 1998, with positive effects of legume presence (Scherer-Lorenzen et al. 2003). LAI and volumetric soil water content were negatively correlated, even at the end of the growing season (October, r = −0·52, P < 0·001; data not shown), implying diversity effects on microclimate through changes in evapotranspiration. Given the local precipitation regime and the well-drained soils, this would imply that the decomposition microenvironment would be rather less favourable – that is, drier – in plots with high biomass and LAI, that is, in high diverse and legume containing communities. But in contrast to this expectation, these plots generally had higher decomposition rates. However, 6 out of 10 weeks of the decomposition experiments were conducted after biomass harvests in autumn when differences in LAI and soil moisture among plots were only minimal, so that the contribution of microclimatic differences among communities might have been of limited importance, supporting the results of Spehn et al. (2000). At sites with less favourable growing conditions, however, decomposition might be stronger controlled by microclimatic effects that are modulated by plant diversity. For example, short-term decomposition of cotton increased with increasing species richness at the dry Mediterranean BIODEPTH site in Greece but decreased at the cool Boreal site in Sweden (Spehn et al. 2005). At both sites, transmittance of light decreased with more species being present. This presumably led to more favourable microenvironmental conditions at the Mediterranean site (higher soil moisture through shading), but to less favourable conditions at the northern site (reduced soil temperatures through shading).

consequences for long-term ecosystem functioning

The results found within this study and those reported by other authors (Naeem et al. 1994; Wardle & Nicholson 1996; Wardle et al. 1997; Mulder et al. 1999; Hector et al. 2000; Spehn et al. 2000; Wardle, Bonner & Barker 2000; Knops et al. 2001; Spehn et al. 2005) suggest that decomposition may be less sensitive to changes in plant diversity than other ecosystem processes like primary productivity. Potential causes of such differences between above- and below-ground processes have been discussed elsewhere (Hooper et al. 2000; Mikola et al. 2002; Naeem 2002; Joshi et al. 2004; Spehn et al. 2005). In essence, soil organisms seem to be functionally rather redundant, as shown by manipulation experiments of soil organisms (Wardle 2002, but see Heemsbergen et al. 2004), and their activity might be more tightly coupled to abiotic conditions than to plant species richness. Additionally, strong control of soil processes by multi-trophic interactions may complicate biodiversity–ecosystem functioning relationships.

Since maintenance of primary production is dependent on the replenishment of a pool of available nutrients, decomposition and subsequent mineralization of detritus are critical for ecosystem sustainability (Swift et al. 1979). Plants can exert strong control on nutrient availability, both directly through uptake, use and loss of nutrients, and indirectly by influencing the activity of soil organisms and herbivory (Hobbie 1992). These effects are ultimately determined by ecophysiological traits of the plants, such as photosynthetic rates, nutrient use and leaf properties. Simultaneously, decomposition processes have large effects on soil fertility and plant growth. Hence, both the producer and decomposer subsystems are obligately dependent on each other and effects of plants on the soil system and vice versa may amplify each other (Wardle 2002). For example, in low-nutrient ecosystems, plants usually grow slowly, allocate more carbon to secondary metabolites and use nutrients efficiently, resulting in low quality litter with low nutrient concentrations, which leads to slow decomposition, leading to even lower nutrient availability. In contrast, on fertile soils, plants grow rapidly, use nutrients inefficiently and hence produce high quality litter which is rapidly decomposed, further increasing nutrient availability and productivity (Hobbie 1992; Wardle 2002). In addition, there is accumulating evidence that plant species select for decomposer communities that preferentially break down the litter from that species, further shaping the contrasting nutrient cycling patterns (Wardle 2002). While these patterns of vegetation–soil interactions have mainly described for widely differing ecosystems (e.g. tundra vs. grasslands), the results from this study imply that such differences may also occur within one ecosystem type (mesic grasslands in this case), depending on functional diversity. Despite having had the same initial nutrient availability on a homogeneous soil, after 2–3 years the experimental communities differed widely in productivity (Hector et al. 1999; Spehn et al. 2005), biomass nitrogen (Spehn et al. 2002), nutrient availability (Gastine et al. 2003; Scherer-Lorenzen et al. 2003; Spehn et al. 2005) and decomposition processes (this study), depending on plant diversity and species composition, with the presence of legumes being especially important. Combining these results, low diverse communities lacking legumes seem to follow more closely the ‘low-nutrient’ trajectory described above: they are less productive, have lower nitrogen concentrations in above-ground biomass and lower decomposition rates, and have lower soil nitrogen availability. High diverse communities with legumes, in contrast, obviously follow the ‘high-nutrient’ option, being more productive, and having higher biomass nitrogen concentrations and higher decomposition rates. However, soil nutrient availability was even reduced at high diversities since available nutrients were efficiently taken up by complementary species, also promoting high productivity (Scherer-Lorenzen et al. 2003). In the long-term, further development of such feedback loops between the producer and decomposer subsystem would imply that the diversity–productivity relationship may become stronger over time, which is indeed supported by the still scarce long-term biodiversity experiments (Tilman et al. 2001; Spehn et al. 2005; van Ruijven & Berendse 2005).


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The field work and parts of the analyses were done during my PhD, supervised by Ernst-Detlef Schulze, to whom I thank for his continuous support and encouragement. Alexandra Prinz, as essential part of the German BIODEPTH team, helped to set up the experiment and to take the numerous measurements. I also thank the whole BIODEPTH team for the spirit of friendship and cooperation, with special thanks to John Lawton and Andy Hector. The experiment was done at the ‘Umweltschutzinformationszentrum Lindenhof, Landesbund für Vogelschutz in Bayern e.V.’ and I am grateful to the staff for the possibility to establish the experiment and for the good cooperation. Jens Schumacher provided the R-code for the calculation of Q and helped with interpretation. I also acknowledge comments provided by Cecilia Palmborg, Eva Spehn and two anonymous reviewers that helped to improve this manuscript.

The BIODEPTH project run from 1996 to 1998 and was funded by the European Commission within the Framework IV Environment and Climate programme (ENV-CT95-0008).


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Balvanera, P., Pfisterer, A.B., Buchmann, N., He, J.-S., Nakashizuka, T., Raffaelli, D. & Schmid, B. (2006) Quantifying the evidence for biodiversity effects on ecosystem functioning and services. Ecology Letters, 9, 11461156.
  • Bardgett, R.D. & Shine, A. (1999) Linkages between plant litter diversity, soil microbial biomass and ecosystem function in temperate grasslands. Soil Biology and Biochemistry, 31, 317321.
  • Cadisch, G. & Giller, K.E. (1997) Driven by Nature: Plant Litter Quality and Decomposition. CAB International, Wallingford.
  • Cornelissen, J.H.C. (1996) An experimental comparison of leaf decomposition rates in a wide range of temperate plant species and types. Journal of Ecology, 84, 573582.
  • Cornelissen, J.H.C. & Thompson, K. (1997) Functional leaf attributes predict litter decomposition rate in herbaceous plants. New Phytologist, 135, 109114.
  • Díaz, S. & Cabido, M. (2001) Vive la différence: plant functional diversity matters to ecosystem processes. Trends in Ecology and Evolution, 16, 646655.
  • Gartner, T.B. & Cardon, Z.G. (2004) Decomposition dynamics in mixed-species leaf litter. Oikos, 104, 230246.
  • Gastine, A., Scherer-Lorenzen, M. & Leadley, P.W. (2003) No consistent effect of plant diversity on root biomass, soil biota and soil abiotic conditions in temperate grassland communities. Applied Soil Ecology, 24, 101111.
  • Hättenschwiler, S. & Vitousek, P.M. (2000) The role of polyphenols in terrestrial ecosystem nutrient cycling. Trends in Ecology and Evolution, 15, 238243.
  • Hättenschwiler, S., Tiunov, A.V. & Scheu, S. (2005) Biodiversity and litter decomposition in terrestrial ecosystems. Annual Review of Ecology, Evolution, and Systematics, 36, 191218.
  • Hector, A., Beale, A.J., Minns, A., Otway, S.L. & Lawton, J.H. (2000) Consequences of the reduction of plant diversity for litter decomposition: effects through litter quality and microenvironment. Oikos, 90, 357371.
  • Hector, A., Schmid, B., Beierkuhnlein, C., Caldeira, M.C., Diemer, M., Dimitrakopoulos, P.G., Finn, J., Freitas, H., Giller, P.S., Good, J., Harris, R., Högberg, P., Huss-Danell, K., Joshi, J., Jumpponen, A., Körner, C., Leadley, P.W., Loreau, M., Minns, A., Mulder, C.P.H., O'donovan, G., Otway, S.J., Pereira, J.S., Prinz, A., Read, D.J., Scherer-Lorenzen, M., Schulze, E.-D., Siamantziouras, A.-S.D., Spehn, E., Terry, A.C., Troumbis, A.Y., Woodward, F.I., Yachi, S. & Lawton, J.H. (1999) Plant diversity and productivity experiments in European grasslands. Science, 286, 11231127.
  • Heemsbergen, D.A., Berg, M.P., Loreau, M., Van Haj, J.R., Faber, J.H. & Verhoef, H.A. (2004) Biodiversity effects on soil processes explained by interspecific functional dissimilarity. Science, 306, 10191020.
  • Hobbie, S.E. (1992) Effects of plant species on nutrient cycling. Trends in Ecology and Evolution, 7, 336339.
  • Hooper, D.U., Bignell, D.E., Brown, V.K., Brussaard, L., Dangerfield, J.M., Wall, D.H., Wardle, D.A., Coleman, D.C., Giller, K.E., Lavelle, P., Van Der Putten, W.H., De Ruiter, P.C., Rusek, J., Silver, W.L., Tiedje, J.M. & Wolters, V. (2000) Interactions between aboveground and belowground biodiversity in terrestrial ecosystems: patterns, mechanisms, and feedbacks. BioScience, 50, 10491061.
  • Hooper, D.U., Chapin, F.S.I., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H., Lodge, D., Loreau, M., Naeem, S., Schmid, B., Setälä, H., Symstad, A.J., Vandermeer, J. & Wardle, D.A. (2005) Effects of biodiversity on ecosystem functioning: a consensus of current knowledge and needs for future research. Ecological Monographs, 75, 336.
  • Hooper, D.U., Solan, M., Symstad, A., Díaz, S., Gessner, M.O., Buchmann, N., Degrange, V., Grime, P., Hulot, F., Mermillod-Blondin, F., Roy, J., Spehn, E.M. & Van Peer, L. (2002) Species diversity, functional diversity, and ecosystem functioning. Biodiversity and Ecosystem Functioning. Synthesis and Perspectives (eds M. Loreau, S. Naeem & P. Inchausti), pp. 195208. Oxford University Press, Oxford, New York.
  • Huston, M.A. (1997) Hidden treatments in ecological experiments: re-evaluating the ecosystem function of biodiversity. Oecologia, 110, 449460.
  • Joshi, J., Otway, S.J., Koricheva, J., Pfisterer, A.B., Alphei, J., Roy, B.A., Scherer-Lorenzen, M., Schmid, B., Spehn, E. & Hector, A. (2004) Bottom-up effects and feedbacks in simple and diverse experimental grassland communities. Insects and Ecosystem Function (eds W.W.Weisser & E.Siemann), pp. 115134. Springer, Berlin, Heidelberg, New York.
  • Knops, J.M.H., Wedin, D. & Tilman, D. (2001) Biodiversity and decomposition in experimental grassland ecosystems. Oecologia, 126, 429433.
  • Lavorel, S. & Garnier, E. (2002) Predicting changes in community composition and ecosystem functioning from plant traits: revisiting the Holy Grail. Functional Ecology, 16, 545556.
  • Mikola, J., Bardgett, R.D. & Hedlund, K. (2002) Biodiversity, ecosystem functioning and soil decomposer food webs. Biodiversity and Ecosystem Functioning: Synthesis and Perspectives (eds M. Loreau, S. Naeem & P. Inchausti), pp. 169180. Oxford University Press, Oxford, New York.
  • Mulder, C.P.H., Koricheva, J., Huss-Danell, K., Högberg, P. & Joshi, J. (1999) Insects affect relationships between plant species richness and ecosystem processes. Ecology Letters, 2, 237246.
  • Naeem, S. (2002) Autotrophic–heterotrophic interactions and their impacts on biodiversity and ecosystem functioning. The Functional Consequences of Biodiversity. Empirical Progress and Theoretical Extensions (eds A.Kinzig, S.Pacala & D.Tilman), pp. 96119. Princeton University Press, Princeton, Oxford.
  • Naeem, S., Thompson, L.J., Lawler, S.P., Lawton, J.H. & Woodfin, R.M. (1994) Declining biodiversity can alter the performance of ecosystems. Nature, 368, 734737.
  • Petchey, O.L. & Gaston, K.J. (2006) Functional diversity: back to basics and looking forward. Ecology Letters, 9, 741758.
  • Prescott, C.E. (2005) Decomposition and mineralization of nutrients from litter and humus. Nutrient Acquisition by Plants. An Ecological Perspective (ed. H.Bassirirad), pp. 1541. Springer, Berlin, Heidelberg, New York.
  • Ricotta, C. (2005) A note on functional diversity measures. Basic and Applied Ecology, 6, 479486.
  • Van Ruijven, J. & Berendse, F. (2005) Diversity–productivity relationships: Initial effects, long-term patterns, and underlying mechanisms. Proceedings of the National Academy of Sciences of the United States of America, 102, 695700.
  • Scherer-Lorenzen, M., Bonilla, J.-L. & Potvin, C. (2007) Tree species richness affects litter production and decomposition rates in a tropical biodiversity experiment. Oikos, 116, 21082124.
  • Scherer-Lorenzen, M., Palmborg, C., Prinz, A. & Schulze, E.D. (2003) The role of plant diversity and composition for nitrate leaching in grasslands. Ecology, 84, 15391552.
  • Schmid, B., Hector, A., Huston, M.A., Inchausti, P., Nijs, I., Leadley, P.W. & Tilman, D. (2002) The design and analysis of biodiversity experiments. Biodiversity and Ecosystem Functioning: Synthesis and Perspectives (eds M.Loreau, S.Naeem & P.Inchausti), pp. 6175. Oxford University Press, Oxford, New York.
  • Seastadt, T.R. (1984) The role of microarthropods in decomposition and mineralisation processes. Annual Review in Entomology, 29, 2546.
  • Spehn, E.M., Hector, A., Joshi, J., Scherer-Lorenzen, M., Schmid, B., Bazeley-White, E., Beierkuhnlein, C., Caldeira, M.C., Diemer, M., Dimitrakopoulos, P.G., Finn, J., Freitas, H., Giller, P.S., Good, J., Harris, R., Högberg, P., Huss-Danell, K., Jumpponen, A., Koricheva, J., Leadley, P.W., Loreau, M., Minns, A., Mulder, C.P.H., O'donovan, G., Otway, S.J., Palmborg, C., Pereira, J.S., Pfisterer, A.B., Prinz, A., Read, D.J., Schulze, E.-D., Siamantziouras, A.-S.D., Terry, A.C., Troumbis, A.Y., Woodward, F.I., Yachi, S. & Lawton, J.H. (2005) Ecosystem effects of the manipulation of plant diversity in European grasslands. Ecological Monographs, 75, 3763.
  • Spehn, E.M., Joshi, J., Alphei, J., Schmid, B. & Körner, C. (2000) Plant diversity effects on soil heterotrophic activity in experimental grassland ecosystems. Plant and Soil, 224, 217230.
  • Spehn, E.M., Scherer-Lorenzen, M., Schmid, B., Hector, A., Caldeira, M.C., Dimitrakopoulos, P.G., Finn, J.A., Jumpponen, A., O'Donnovan, G., Pereira, J.S., Schulze, E.-D., Troumbis, A.Y., & Körner, C. (2002) The role of legumes as a component of biodiversity in a cross-European study of grassland biomass nitrogen. Oikos, 98, 205218.
  • Swift, M.J., Heal, O.W. & Anderson, J.M. (1979) Decomposition in Terrestrial Ecosystems. Blackwell Science, Oxford.
  • Tilman, D. (2001) Functional diversity. Encyclopaedia of Biodiversity (ed. S.A.Levin), pp. 109120. Academic Press, San Diego.
  • Tilman, D., Reich, P.B., Knops, J., Wedin, D., Mielke, T. & Lehmann, C. (2001) Diversity and productivity in a long-term grassland experiment. Science, 294, 843845.
  • Wardle, D.A. (2002) Communities and Ecosystems: Linking the Aboveground and Belowground Components. Princeton University Press, Princeton, Oxford.
  • Wardle, D.A. & Nicholson, K.S. (1996) Synergistic effects of grassland plant species on soil microbial biomass and activity: implications for ecosystem-level effects of enriched plant diversity. Functional Ecology, 10, 410416.
  • Wardle, D.A. & Van Der Putten, W.H. (2002) Biodiversity, ecosystem functioning and above- and below-ground linkages. Biodiversity and Ecosystem Functioning: Synthesis and Perspectives (eds M.Loreau, S.Naeem & P.Inchausti), pp. 155168. Oxford University Press, Oxford, New York.
  • Wardle, D.A., Bonner, K.I. & Barker, G.M. (2000) Stability of ecosystem properties in response to above-ground functional group richness and composition. Oikos, 89, 1123.
  • 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, 247258.