u n i ve r s i t y o f co pe n h ag e n

An isotopic method for testing the influence of leaf litterquality on carbon fluxesduring decomposition

Rubino, Mauro

Published in:Oecologia

DOI:10.1007/s00442-007-0815-y

Publication date:2007

Document versionPublisher's PDF, also known as Version of record

Citation for published version (APA):Rubino, M. (2007). An isotopic method for testing the influence of leaf litterquality on carbon fluxes duringdecomposition. Oecologia. https://doi.org/10.1007/s00442-007-0815-y

Download date: 06. May. 2021

ECOSYSTEM ECOLOGY

An isotopic method for testing the influence of leaf litterquality on carbon fluxes during decomposition

Mauro Rubino Æ C. Lubritto Æ A. D’Onofrio ÆF. Terrasi Æ G. Gleixner Æ M. F. Cotrufo

Received: 13 December 2005 / Accepted: 5 July 2007 / Published online: 31 July 2007

� Springer-Verlag 2007

Abstract During microbial breakdown of leaf litter a

fraction of the C lost by the litter is not released to the

atmosphere as CO2 but remains in the soil as microbial

byproducts. The amount of this fraction and the factors

influencing its size are not yet clearly known. We per-

formed a laboratory experiment to quantify the flow of C

from decaying litter into the soil, by means of stable C

isotopes, and tested its dependence on litter chemical

properties. Three sets of 13C-depleted leaf litter (Liquid-

ambar styraciflua L., Cercis canadensis L. and Pinus taeda

L.) were incubated in the laboratory in jars containing 13C-

enriched soil (i.e. formed C4 vegetation). Four jars con-

taining soil only were used as a control. Litter chemical

properties were measured using thermogravimetry (Tg) and

pyrolysis–gas chromatography/mass spectrometry–com-

bustion interface–isotope ratio mass spectrometry (Py–GC/

MS–C–IRMS). The respiration rates and the d13C of the

respired CO2 were measured at regular intervals. After

8 months of incubation, soils incubated with both L. sty-

raciflua and C. canadensis showed a significant change in

d13C (d13Cfinal = �20.2 ± 0.4% and �19.5 ± 0.5%,

respectively) with respect to the initial value (d13Cinitial =

�17.7 ± 0.3%); the same did not hold for soil incubated

with P. taeda (d13Cfinal:�18.1 ± 0.5%). The percentages

of litter-derived C in soil over the total C loss were not

statistically different from one litter species to another.

This suggests that there is no dependence of the percentage

of C input into the soil (over the total C loss) on litter

quality and that the fractional loss of leaf litter C is

dependent only on the microbial assimilation efficiency.

The percentage of litter-derived C in soil was estimated

to be 13 ± 3% of total C loss.

Keywords Pyrolysis–gas chromatography/mass

spectrometry–combustion interface–isotope ratio mass

spectrometry � Respiration rates � Soil carbon input �Stable carbon isotopes � Thermogravimetry

Introduction

Litter decomposition is defined as the breakdown of

organic macromolecules from dead plant/animal material

into inorganic constituents (such as CO2, water and inor-

ganic nutrients). However, a set of organic compounds of

smaller molecular weight, often much more resistant to

microbial breakdown than the initial material, are also

formed during the process. These compounds significantly

contribute to the build up of organic matter in soils (Zech

and Kogel-Knaber 1994). Yet their rate of incorporation

into the soil organic C pool in relation to the rates of litter

decay has been largely ignored, and attention has focused

primarily on quantifying the rate of mass loss or of CO2

efflux from litter (Cortez and Bouche 2001; Cortez et al.

1996; Gartner and Cardon 2004). However, it may be

Communicated by Dan Yakir.

M. Rubino � C. Lubritto � A. D’Onofrio � F. Terrasi �M. F. Cotrufo

Dipartimento di Scienze Ambientali,

Seconda Universita di Napoli,

Via Vivaldi 43, 81100 Caserta, Italy

G. Gleixner

Max Planck Institute for Biogeochemistry,

Winzerlaer Strasse 10, 07745 Jena, Germany

M. Rubino (&)

Unite UR1119 Unite Geochimie des Sols et des Eaux,

INRA, 13545 Aix en Provence, France

e-mail: mauro.rubino@unina2.it

123

Oecologia (2007) 154:155–166

DOI 10.1007/s00442-007-0815-y

hypothesised that the rate of incorporation of C derived

from above-ground litter decay into soil organic matter

(SOM) is a function of the litter decay rates and microbial

use efficiency (defined as the ratio of C incorporated in the

microbial biomass to the C respired as CO2). For this

reason, it may be controlled by the same factors that con-

trol litter decomposition.

Litter decomposition rate is determined by the prevail-

ing environmental conditions for a given location, by the

intrinsic chemical properties of plant litter (often referred

to as ‘‘litter quality’’) and by the diversity and activity of

the decomposer community (Swift et al. 1979). With

respect to litter quality, differences in decomposition rates

are a consequence of the type and relative concentration of

chemical compounds present in litter. According to Chapin

et al. (2002), these compounds can be broadly categorized

as: labile metabolic compounds (such as sugars and amino

acids), moderately labile structural compounds (such as

cellulose and hemicellulose) and recalcitrant structural

material (such as lignin and cutin). Although the chemical

characterization of litter gives important information on its

degradability, there is no universally accepted litter quality

index. The ratio of C to N concentration (C/N ratio), for

example, has frequently been used as an index of litter

quality (Enriquez et al. 1993; Gholz et al. 2000). In

recalcitrant litter, the concentration of lignin or the lignin/N

ratio is often a good predictor of the decomposition rate

(Berg and Staaf 1980; Melillo et al. 1982; Taylor et al.

1989). Useful quality indices may be derived by indirect

methods, such as near-infrared spectroscopy, which has

been shown to be a reliable indicator of decay rates (Gillon

et al. 1999; Joffre et al. 2001; Perez-Harguindeguy et al.

2000).

The current inability to unequivocally relate litter

quality to decay rates is reflected in models. Most models

represent litter by two or more discrete pools, and criteria

of compartmentalisation are defined either chemically

(Minderman 1968), kinetically (Jenkinson et al. 1991) or

functionally (Parton et al. 1994). In any case, litter

decomposition models are presently unable to describe the

influence of litter quality and its control on C transfers to

SOM (Corbeels 2001).

Our understanding of litter decomposition is currently

constrained by the methods used to quantify it. In field

experiments, decomposition is commonly measured as the

loss of mass from a known amount of litter, using the litter

bag method (Berg and Ekhbom 1991; Bocock and Gilbert

1957; Falconer et al. 1933). Decomposition curves derived

from litter bag studies performed on a time scale of several

years are best described by the presence of an asymptotic

value, which may represent the mass of undecomposed

litter that gets incorporated by SOM as litter debris,

although this is yet to be proved. In laboratory studies,

decay rates are generally obtained by measuring the CO2

evolved from litter debris due to microbial activity

(Anderson and Ineson 1982). Occasionally, dissolved

organic C leached from the decomposing system has been

determined (Cotrufo et al. 1994). The use of classical

methodologies makes the quantification of C input from

decomposing litter to the soil very difficult, also because

the contribution of litter-derived C to the SOM is small

compared to the total soil organic C content, and conven-

tional methods of quantifying C changes in soils are not

sensitive enough to allow accurate determination of it

(Hungate et al. 1996; Liski et al. 1998). However, the

quantification of the input of organic C from decaying litter

into SOM is crucial since the leaf litter pool is an important

fraction of the total C in terrestrial ecosystems (Matthews

1997).

Stable isotope methodologies have proved to be useful

tools in quantifying C fluxes to and from different C pools.

Provided that two C sources have a sufficiently different

stable isotope composition, it is possible to quantify their

relative contribution to a third C pool resulting from the

mixing of the two, by applying a two-source mixing model

(Balesdent et al. 1987; Leavitt et al. 2001; Lichter et al.

2005).

This method can be used to quantify litter-derived C in

SOM, providing that litter and soil have a different isotopic

composition.

The aim of this study was to quantify the relative

proportion of the two different fractions of C lost from the

decomposing litter (i.e. the one which enters the soil and

the one released as CO2 to the atmosphere) by using an

isotopic mass balance approach and to determine if litter

quality influences the relative proportion of C incorporated

into the soil and respired as CO2. In other words, we

hypothesize that the quantity of C that is lost from leaf

litter pools and transferred to the SOM pool, during

decomposition, is not proportional to the total C loss.

Moreover, since lignin is cometabolically broken down,

and during its decay CO2 is produced with no C gain for

the micro-organisms (Fritsche 1998), we further hypothe-

size that the higher the lignin concentration of the litter,

the higher is the fraction of respired CO2 over the total C

loss.

To test these hypotheses, 13C-depleted leaf litter, from

three woody species known to be different for litter quality

and decomposability (Cotrufo et al. 2005), were incubated

on the surface of a 13C-enriched soil, for 8 months under

laboratory controlled conditions. At the end of the exper-

iment, the change in the soil 13C enrichment was measured

and the isotopic mass balance applied to quantify C

incorporation derived from the litter. Different chemical

properties of litters are quantified and related to differences

in litter C dynamics.

156 Oecologia (2007) 154:155–166

123

Materials and methods

Samples

Leaf litter from three different woody species (two broad-

leaved, Liquidambar styraciflua and Cercis canadensis,

and one coniferous, Pinus taeda) derived from the first

forest–atmosphere C transfer and storage (FACTSI)

experiment (Hendrey et al. 1993) at Duke forest (Duke

University, Durham, N.C.), were used in this experiment

because of their very high 13C depletion, resulting from

exposure to an atmosphere enriched with fossil fuel-

derived CO2 (Table 1). Conversely, a 13C-rich soil

(Table 1) was obtained from a long-term maize crop

located in north-eastern Italy (Flaibano 46�3034N,

12�5908E, 77 m asl). The soil is classified as an alluvial

mesic Udifluvent (USDA classification). The soil and litter

materials had already been used for previous experiments

(see, Cotrufo et al. 2005; Del Galdo et al. 2003), thus their

C properties were well known and considered suitable for

the purposes of this work (Table 1). In fact, given the large

variation in the natural abundance of their stable C iso-

topes, even a small amount of C input deriving from the

litter was anticipated to cause a detectable variation in the

soil isotopic composition.

Preparation of experimental units

Litter samples were collected at the FACTSI experimental

site in October 1998 (Cotrufo et al. 2005), air-dried and

stored in paper bags. In June 2003 they were rewetted (13 g

distilled water for 3 g dry litter) and incubated in labora-

tory units on the top of soil. On the day of unit preparation,

a sub-sample of each litter (1 g) was oven dried (at 70�C

for 48 h) for dry weight correction. The soil was collected

at the Flaibano site in July 2000 as described by Del Galdo

et al. (2003), treated with HCl to remove carbonates (Harris

et al. 2001), sieved to 2 mm, well mixed to homogenise the

organic C content, air dried and stored, prior to use for this

experiment. Sub-samples of 70 g soil were added to each

incubation unit, consisting of 1.5-l air-tight glass jars. Soil

was rewetted to 55% of water holding capacity using the

water resulting from the litter hydration process, and

inoculated with a fresh soil suspension (3 ml, 1:3 water–

organic soil). Litter samples (3.0 ± 0.1 g) of each species

were added to the incubation units containing soil samples,

with four replicate units per litter species. Additionally four

replicates of only soil were used as control systems,

resulting in a total of 16 experimental units (i.e. three litter

species + soil and one soil only, all replicated 4 times). The

jars were incubated closed, in the dark, at 20�C, in a

temperature-controlled cabinet.

Respiration measurements and harvest

For the entire duration of the experiment, on a daily basis,

the jars were opened and flushed with room air to maintain

aerobic conditions. Once a week jars were weighed to

determine eventual water loss. If needed, they were rew-

etted to the initial weight, to keep soil moisture constant.

To measure the d13C of the CO2 respired by the system,

at regular intervals, jars were flushed with CO2-free air (8 l

air, to eliminate the contamination from atmospheric CO2)

and kept closed for 2 days, to allow the respired CO2 to

accumulate in the jar head-space. The respired CO2 was

then trapped by means of a cryogenic process (Bertolini

et al. 2005). To connect the jars to the cryogenic line, two

holes were made on the jar cap; 6-mm-diameter teflon

tubes were fitted into the holes: the inlet tube connected to

a container filled with soda lime (to avoid CO2 entering the

Table 1 Chemical characteristics [d13C(Vienna Pee Dee belemnite; VSPDB),

C and N concentrations and C/N ratio] of litter and soil samples

before and after the 8-month incubation period. Numbers in

parentheses are SDs (n = 8 instrumental replicates for the initial

samples; n = 4 experimental replicate for the samples after

incubation)

Samples types Before incubation After incubation

d13CVSPDB (%) C (%) N (%) C/N ratio d13CVSPDB (%) C (%) N (%) C/N ratio

Litter samples

Liquidambar styraciflua �43.0 (0.2) 46.37 (0.02) 0.72 (0.02) 64 (2) �43.4 (0.9) 45 (4) 1.02 (0.07) 44 (5)

Cercis canadensis �43.7 (0.3) 43.9 (0.3) 1.24 (0.03) 35 (1) �43.9 (0.3) 44 (2) 1.39 (0.14) 32 (3)

Pinus taeda �43.7 (0.7) 49.3 (0.5) 0.590 (0.004) 84 (1) �43.3 (0.5) 52 (2) 0.66 (0.02) 78 (4)

Soil samples

Control �17.7 (0.3) 0.87 (0.01) 0.105 (0.002) 8.3 (0.2) �17.9 (0.3) 0.86 (0.06) 0.109 (0.003) 7.9 (0.5)

L. styraciflua �20.2 (0.4) 1.02 (0.10) 0.107 (0.005) 9.5 (1.0)

C. canadensis �19.5 (0.5) 0.98 (0.06) 0.111 (0.006) 8.7 (0.7)

P. taeda �18.1 (0.5) 0.94 (0.03) 0.109 (0.003) 8.6 (0.4)

Oecologia (2007) 154:155–166 157

123

jars) and the outlet tube connected to the cryogenic line.

Each cryogenic process lasted 20 min, with a flow rate of

the gas entering the line of 0.1 l/min. After the cryogenic

process, jars were left open for half an hour in order to

equilibrate with the atmospheric CO2 concentration within

a thermostatic bath at 20�C. Following equilibration, jars

were closed, connected to an infrared gas analyzer (IRGA;

ADC 2250 Series, Bioscientific) and the respiration rate

was measured in a closed dynamic mode. During a mea-

surement period, lasting 30 min, the CO2 concentration

was monitored with a frequency of 1 Hz and respiration

rate obtained by interpolating the experimental values with

a linear fit in a CO2 concentration versus time graph (mean

r2 = 0.999). After 8 months, surface litter was removed

from each jar, carefully cleaned of soil, oven dried at 70�C

and weighed for determination of mass loss.

Chemical and isotopic analyses

Each litter type before and after the incubation was

analysed for its chemical properties. Approximately

10 mg of each litter sample (i.e. a sub-sample of each

litter species before incubation and of each replicate after

incubation) was analysed by thermogravimetry (Tg) (Pope

and Judd 1977) to measure the degradability at increasing

temperatures. At the same time, ca. 1 mg of each litter

type (in two replicates) was analysed by pyrolysis–gas

chromatography/mass spectrometry–combustion interface–

isotope ratio mass spectrometry (Py–GC/MS–C–IRMS)

(Gleixner et al. 1999; Gleixner and Schmidt 1998) in order

to quantify the percentage of different groups of C-based

compounds. Chemical analyses of litter were performed at

Max Planck Institute for Biogeochemistry (Jena,

Germany).

Soil samples were air dried, milled and analysed for

their C, N and 13C content. Isotopic composition

[d13CVienna Pee Dee belemnite (VSPDB) (%)] and C and N

concentrations (%) for soil and litter samples before and

after the incubation period were measured using an ele-

mental analyzer (ThermoFinnigan EA 1112) coupled with

an IRMS (ThermoFinnigan Deltaplus) at the Department of

Environmental Science (Second University of Naples,

Caserta, Italy).

Data analysis

The determination of the fraction of C loss from litter and

introduced into the soil was obtained from the variation in

d13C of soil and litter and by applying a two-source mixing

model (Balesdent et al. 1987):

Cls tð Þ

Cs tð Þ ¼ds tð Þ � dc tð Þdl tð Þ � dc tð Þ

whereCl

sðtÞCsðtÞis the fraction of litter-derived C in soil (Cl

sðtÞ,s = soil, l = litter) over the total soil C, ds(t) is the d13C of

soil at the end of incubation, dc(t) is the d13C of control soil

at the end of incubation (c = control) and dl(t) is the d13C

of the litter sample. The main assumption with this mass

balance is that the d13C of litter-derived C compounds in

soil is equal to the d13C of the bulk litter.

The determination of the litter- and soil-derived CO2

was obtained by applying an isotopic mass balance to the

total CO2 respired by litter and soil together:

FlðtÞFðtÞ ¼

dFðtÞ � dFcðtÞdF1ðtÞ � dFcðtÞ

whereFlðtÞFðtÞ is the fraction of litter-derived CO2 [Fl(t)] over

the total CO2 respired by the system (soil and litter) [F(t)],

dF(t) is the d13C of the total CO2 respired by the system,

dFc(t) is the d13C of CO2 respired by the control soil and

dFl(t) is the d13C of litter sample.

Main assumptions with this mass balance are:

1. d13C of CO2 respired by the soil in the mixed system is

equal to d13C of CO2 respired by the control soil.

2. d13C of CO2 respired by litter in the mixed system is

equal to d13C of solid litter (no fractionation during the

respiration process).

The consequences of these assumptions will be discussed

in the ‘‘Discussion’’. Using the results of this mass balance

and the data on respiration rate over time, the total CO2

respired by the litter was obtained by numerical

integration.

Decay constants for each litter species were obtained by

interpolating the measured respiration rates over time.

Curve fitting used a double exponential regression

y ¼ a1e�k1t þ a2e�k2t� �

; where k1 is the decay constant for

the early stages of the decomposition process (i.e. when

mostly labile compounds in litter are respired), whereas k2

is the decay constant for the late phase of the decomposi-

tion process (i.e. when mostly recalcitrant compounds are

respired).

t-tests were applied every time to compare the differ-

ence between the same variable measured in two different

litter species or two different periods (i.e. beginning and

end of incubation). When significance was found, the sig-

nificance level is indicated in the text in parenthesis

(P < 0.05 or P < 0.01). Regarding functional relations

between initial litter quality and C losses, since only three

litter species were investigated, they cannot be tested by r.

However, the significance of regressions could be checked

for the 12 replicates at the end of incubation, considering

158 Oecologia (2007) 154:155–166

123

each replicate as an independent sample. The significance

level of the regression and r are given in parenthesis. One

of the experimental units (a replicate containing L. sty-

raciflua) was proven to be unreliable and was not used in

the statistical analyses.

Results

Chemical properties of litters

Thermogravimetry

Tg was applied to see how the different litter species,

before and after the incubation period, degraded during a

pyrolysis-combustion process. As an example, Fig. 1 rep-

resents the trend of mass loss for L. styraciflua before the

incubation. Not considering the first part (20–200�C: water

loss) and the last one (850–20�C: mass stabilization and

ash residue) the trend can be divided into three steps,

within which the slope can be considered constant:

1. Rapid mass loss in the temperature range 200–400�C

under Ar flow.

2. Slow mass loss in the temperature range 400–850�C

under Ar flow.

3. Sudden mass loss at 850�C when O2 is injected.

The chemical characteristics of each litter species before

and after the incubation period can be compared by

comparing the mass loss in these three intervals (in the

following referred to as steps 1, 2 and 3). In Table 2 the

percentages of mass loss are shown. Among the three litter

species before the incubation period, P. taeda showed the

highest percentages of mass loss in steps 1 and 2 and the

lowest in step 3 (P < 0.01). Regarding differences in and

between litters prior to and after the incubation period, a

significant change (P < 0.05) in the percentage of mass

loss in steps 1 and 2 was registered for L. styraciflua; an

increase (P < 0.01) in the percentage of mass loss in step 2

was the only significant variation for C. canadensis; and a

significant decrease (P < 0.01) in steps 1 and 2 and an

increase in step 3 were measured for P. taeda.

In order to have a rapid way to compare the character-

istics of litters and to summarize the indications of the Tg

results, a quality index (Table 2) was calculated as the ratio

of the mass loss during the pyrolysis steps (1 + 2) to the

mass loss during the combustion 3; Q ¼ ð1þ2Þ3

� �: In prin-

ciple, the lower the quality index the higher the quantity of

recalcitrant compounds in the litter. Correspondingly, a

litter with a higher quality index should show higher con-

centrations of easily degraded C compounds. Among litter

species before the incubation period, P. taeda had the

highest quality index and L. styraciflua the lowest

(P < 0.01). P. taeda and C canadensis show a significant

decrease (P < 0.01) in the quality index with incubation.

Pyrolysis–gas chromatography/mass spectrometry–

combustion interface–isotope ratio mass spectrometry

Py-GC/MS-C-IRMS measures the relative proportion of

different compounds produced during a Curie-point (500�C)

pyrolysis process. These compounds can be divided into

different classes according to the molecule each compound

was generated from. In Fig. 2 the percentages of compounds

derived from: carbohydrates, lignin, lipids and others (i.e.

proteins, phenols, cycloalkanes and benzenes) are shown for

each litter type. The left-hand panel of Fig. 2 refers to the

litter before the incubation period, whereas the right-hand

one refers to the litter after the incubation period. Among the

three species before the incubation period, P. taeda had the

highest percentages in lignin and lipids and the lowest in

carbohydrates and others compounds (P < 0.01). By com-

paring litters before and after the incubation (comparing

Fig. 2 left-hand and right-hand panels) a significant

enrichment in carbohydrates (P < 0.01) and depletion in

lignin (P < 0.05) was found for L. styraciflua. No significant

change in carbohydrates and lipids was measured for C.

canadensis. A decrease in the percentage of lipids was the

only significant variation for P. taeda (P < 0.05).

Respiration measurements and d13C of respired CO2

For all systems (litter and soil), respiration rates decreased

over time and became relatively constant after 100 days of

0

20

40

60

80

100

0 1000 2000 3000 4000 5000 6000 7000 8000

Time (s)

fo noitcarf gniniame

R)

%( ssam laitini

0

200

400

600

800

1000

)C°( erutarep

meT

Mass

Temperature

Pyrolysis Combustion

O2 injection

1

2

3

Pyrolysis

Fig. 1 Mass remaining (% of initial mass, solid line) and temperature

variation with time (�C, dotted line) in the thermogravimetry process

for Liquidambar styraciflua at the beginning of the experiment. The

three steps described in the text are indicated as mean slopes for the

curve. Moreover, the pyrolysis and combustion periods are shown in

the upper part of the figure together with the point at which O2 was

injected

Oecologia (2007) 154:155–166 159

123

incubation. Respiration rates for L. styraciflua and C.

canadensis systems (respiration rates of litter and soil

together) were not statistically different (P < 0.01), with

the exception of the third and the forth sampling days

(Fig. 3). Respiration rates for P. taeda systems were

always significantly lower than those of the other litters

(P < 0.01). As expected, the respiration rate for control soil

was nearly constant during the entire incubation period

(with the exception of the first sampling day, when the

higher value was likely a result of unit preparation and soil

biomass reactivation).

Figure 3 also shows the d13CVSPDB of the CO2 respired

by each litter and soil. Not considering the first sampling

day (d13C higher than in the following period for all litters

and soils, most likely another effect of disturbance due to

unit preparation), no trend was evident. On the contrary,

the d13C of the CO2 respired from control soil showed a

well-defined trend over time towards more negative values

(from *�16 to �22%). Qualitatively, data of respiration

rates and isotopic composition were in agreement. In fact,

assuming a constant contribution to respiration from soil,

the higher the CO2 respired by litters (the higher the

measured respiration rates) the closer the d13C(CO2) was to

the d13C of litters.

By weighting the respiration rates with the d13C value of

the respired CO2 (as explained in the data analysis), the

litter derived C–CO2 can be separated from the soil-derived

C–CO2. This allows the quantification of the total C–CO2

respired from each litter (Table 3). In Table 3 the decom-

position constants (k1 and k2) together with the asymptotic

coefficients a1 and a2 are also shown. The half-life values

of the recalcitrant compounds for all litters, except C.

canadensis, were much longer than the entire period of

incubation; for this reason very low decomposition con-

stants (k2) for the late stage of decomposition were found

for L. styraciflua and P. taeda. For these two species, this

caused the coefficient a2 to be proportional to the asymp-

totic value that would be measured when mass loss

becomes extremely low in a fit with a single exponential

and a constant.

Table 2 Percentage of mass loss during thermogravimetry analysesa

in different steps (1, 2 and 3) with respect to the total mass loss in the

three steps (see Fig. 1 as an example) and quality index derived from

the losses of mass during the three different steps. Errors (in

parentheses) for samples before the incubation are SDs over three

instrumental replicates; errors for samples after the incubation are

SDs over four experimental replicates

Step number % Mass loss of the total Quality index (%)

L. styraciflua C. canadensis P. taeda L. styraciflua C. canadensis P. taeda

Before the incubation 1 23.2 (0.1) 26.1 (0.1) 29.0 (0.1) 348 (2) 402 (2) 466 (2)

2 2.61 (0.01) 2.54 (0.01) 2.77 (0.01)

3 74.2 (0.2) 71.3 (0.2) 68.2 (0.2)

After the incubation 1 21 (4) 23 (3) 23 (1) 314 (107) 358 (44) 345 (12)

2 3.1 (0.7) 3.2 (0.2) 2.67 (0.05)

3 76 (22) 74 (5) 74 (1)

aThe temperature ranges corresponding to each step are: 320 ? 340�C for step 1, 520 ? 540�C for step 2 and 850�C ? O2 injection for step 3.

The ranges are narrower than those mentioned in the text in order to avoid border effects which differ from one sample to another

After incubation

adeat .Psisnedanac .Caulficaryts .L

Before incubation

0

20

40

60

adeat .Psisnedanac .Caulficaryts .L

Am

ount

of c

ompo

unds

(%) setardyhobraC ningiL

sdipiL srehtO

Fig. 2 Relative proportion (% of total amount of compounds) for different classes of compounds in different litter species before and after the

incubation period. Errors are SDs over three instrumental replicates

160 Oecologia (2007) 154:155–166

123

Soil C and litter derived C input

In Table 1 the d13CVSPDB (%) and C content (%) at the end

of incubation for control soil and soils incubated with each

litter are listed. With respect to the soil values before the

incubation period (d13CVSPDB = �17.7 ± 0.3%, C con-

tent = 0.87 ± 0.01%), no significant change either in

isotopic composition or in C content were measured for the

control soil. On the other hand, significant depletion in 13C

and enrichment in C content was evident for both soils

incubated with L. styraciflua and C. canadensis litters

(P < 0.01). P. taeda had a significant increase in the C

content (P < 0.01) and no change in d13C.

Leaf litter mass loss, expressed in terms of C loss as a

percentage of the initial mass, decreased in the order:

L. styraciflua, C. canadensis, P. taeda (Table 3; P < 0.05).

In comparison to litter samples before the incubation per-

iod, P taeda was the only one to show a significant increase

in C percentage (P < 0.05), while no significant variation

in d13CVSPDB was measured for any litter species. All litters

show a decrease in the C/N ratio (Table 1), which is more

significant for L. styraciflua (P < 0.01) and less significant

for C. canadensis and P. taeda (P < 0.05).

Using the data on variation in d13C of soil, litter mass

loss and litter C content and applying an isotopic mass

balance, the fraction of C loss from litter samples as an

0.000

0.002

0.004

0.006

0.008

OC-

C deripseR

2yad g(

1-)

Control soil

C. canadensis & soil

P. taeda & soil

L. styraciflua & soil

0 50 100 150 200

Time (d)

δ31

OC(

C2)

BD

Psv)

‰(

-45

-40

-35

-30

-25

-20

-15

Fig. 3 Respiration rates (g of C

respired as CO2 per day) for

each system (error bars are SDs

over four replicates) and

isotopic composition

[d13CVSPDB (%)] of CO2

respired by each system (errorbars are SDs over four

experimental replicates)

Table 3 C lost by litter samples. Errors are SDs over four experimental replicates

Type of sample a1 (g days�1)a a2 (g days�1)a k1 (days�1)a k2 (days�1)a Rv2 Total C

loss (g)bCumulative

C–CO2 (%)bC input in

soil (%)b

L. styraciflua 45.4 (1.4 · 10�4) 21.2 (0.6 · 10�4) 0.026 (0.005) <10�5 12 0.50 (0.10) 91 (2) 15 (7)

C. canadensis 28.4 (2.1 · 10�4) 35.2 (0.4 · 10�4) 0.13 (0.02) 4.3 (0.3 · 10�3) 48 0.31 (0.11) 92 (2) 14 (4)

P. taeda 16.7 (0.5 · 10�4) 7.6 (0.2 · 10�4) 0.016 (0.002) <10�5 20 0.18 (0.05) 96 (4) 6 (7)

a Decomposition constants (k1 and k2) and asymptotic values (a1 and a2) derived from a double exponential fit (y ¼ a1e�k1t þ a2e�k2t), goodness

of fit is given by the Rv2 factor of respiration rates in the upper panel of Fig. 3ab Total C lost by each litter; cumulative C–CO2 and C input in soil (both expressed as % of total C loss) of each litter species at the end of the

incubation

Oecologia (2007) 154:155–166 161

123

input to soil C was determined, as explained in the data

analysis; results are presented in Table 3.

Discussion

Chemical properties of litters: Tg versus

Py-GC/MS-C-IRMS

It is generally accepted that litter quality is one of the factors

controlling the rate of litter decay, but, despite the many

quality indexes proposed, there is neither a common defi-

nition nor a widely accepted quantitative index of ‘‘quality’’

(Heal et al. 1997; Joffre et al. 2001). A litter of high quality

is, by definition, a litter with a fast decay rate. Since fast

decay rate implies abundant nutrient and energy supply and

high energy for microbial use, litter quality has been cor-

related to high nutrient concentrations (Berg et al. 1987;

Melin 1930; Millar et al. 1948) and high energy C-based

compounds requiring low activation energy to be degraded

(Berg and Agren 1984; McClaugherty and Berg 1987).

In this study, we applied Tg to derive a quality index

based on thermo-degradability properties of litter

(Table 2); the quality index measured for each litter species

tended to increase as the lipid content in litters increased

and to decrease with the increasing percentage of com-

pounds referred to as ‘‘other’’ from Py-GC/MS-C-IRMS.

Unfortunately, we only had access to three 13C-depleted

litter species to use in this experiment, thus, we could not

test for significant relations between concentrations of C-

based compounds and quality index. Future studies should

address the question using a larger variety of litter species.

It has to be noted that the class of compounds referred to

as ‘‘other’’ is quite heterogeneous, consisting of more

(proteins) and less (phenols) labile molecules. Contrary to

our original hypothesis, the percentage of mass loss in

phase 2 of Tg was found to go in the opposite direction of

the carbohydrate content of the three litters investigated

and to follow the trend of the lipids; similarly phase 1

followed the trend of lignin content in the litter species.

Overall, there was no relationship between the concentra-

tion of labile compounds (given by Py-GC/MS-C-IRMS)

and chemical resistance to degradation using temperature

(given by Tg). This could be due to different mechanisms

of degradation due to microbial and thermal attack. A

possible explanation is that microbial enzymes act as a

catalyst by lowering the binding energy in a biochemical

reaction (Gleixner et al. 2001). On this basis, we do not

advise the use of Tg to produce an integrated index of litter

quality, at least until the correspondence between thermal

and microbial degradation is better understood and the

chemical composition of compounds being combusted at

the different stages identified.

Litter–soil C dynamics during incubation

Total C loss for each litter species tended to increase with

increasing percentages of others (as defined by Py-GC/MS-

C-IRMS), present in each species at the beginning of the

incubation, and to decrease with increasing lipid content at

the beginning of the incubation. As described above, the

group of compounds referred to as ‘‘other’’ is a heteroge-

neous class of molecules, the dependence of the total C loss

on the percentage of ‘‘others’’ would suggest a stronger

influence of metabolic molecules, such as proteins, rather

than structural molecules, such as phenols, on this group of

compounds. The total C loss was positively correlated

(r = 0.854; P < 0.01) to the carbohydrate content and

negatively correlated (r = 0,720; P < 0.05) to the lignin

content, both measured at the end of incubation. This is in

agreement with Taylor et al. (1989) indicating that lignin

control over the decomposition process increases as

decomposition proceeds.

A large variability of results among replicates for each

litter species complicated the interpretation of causal

relations. This variability was found both in chemical

characteristics and in mass losses (e.g. errors for the total C

losses shown in Table 3).

C lost during decomposition: CO2 respired and decay

constants

Our experiment is a short-term decomposition study and

gives us information about the first stages of the leaf litter

decomposition process. Nonetheless, Fig. 3 demonstrates

that litter mass loss had already stabilised after 100 days of

incubation. For this reason, we can assume that the rate of C

loss would not change significantly in the following stages

of decomposition; thus our results should be comparable to

those from longer term decomposition experiments.

In laboratory tests, litter decomposition is considered as

the mineralization of C from litter residue due to microbial

respiration, and thus is measured as CO2 production, with

the evolved CO2 often trapped in NaOH and determined by

titration. This gives a measure of the quantity of CO2

produced over a period of time (that is, a mean value over

that period). In our experiment, the CO2 evolved from the

systems was determined by IRGA at periodic discrete

times. The measurement of the quantity of CO2 ‘‘instan-

taneously’’ evolved from the system under investigation,

which does not correspond to an average over a period of

time, is generally considered a more accurate way to follow

the dynamics of CO2 production over time. These data can

be either averaged or used as instantaneous data.

Unfortunately, our measurements of the amount of CO2

respired show a discrepancy with the total C lost by litters.

162 Oecologia (2007) 154:155–166

123

For each replicate, the C–CO2 lost by litters was higher

than total C loss. We can think of two possible explana-

tions for this inconsistency:

1. A lower estimation of litter mass loss due to soil

contamination of litter debris, or to ingrowth of fungal

mycelia, which induced overestimation of final litter

dry weight. It is not easy to remove soil C completely

from the litter surface because soil tends to adhere

strongly to the litter; furthermore, litter becomes more

and more fragile with time and removing soil from its

surface becomes difficult. Fungal mycelia ingrowth on

litter was certainly evident, and it could not be

removed, although we expected its weight to be minor

compared to that of the litter. Because of this, the litter

mass loss may be underestimated.

2. Overestimation of cumulative C–CO2 evolved from

the litter. When jars are opened, after being closed for

2 days and after the sampling process (by cryogenic

extraction), the rapid increase in the O2 concentration

could stimulate the activity of aerobic micro-organ-

isms causing respiration rates systematically higher

than those during the incubation period.

Nevertheless, as described above, respiration rates show a

regular trend and are in good agreement with data of

isotopic composition of respired CO2. Thus we believe that

a comparison between these two sets of data is still possible.

The total CO2 respired by litter species was found to be

positively correlated to the total C loss (r = 0.807,

P < 0.01). This means that the percentage of C loss as

respired CO2 over the total C loss (measured as the sum of

C–CO2 loss and C loss as an input in soil) did not change

among the different litter species (Table 3). This is a first

indication that the fractions of C–CO2 loss over the total C

loss during litter decomposition is not dependent on litter

quality, but is rather a constant fraction of mass loss.

In our experiment, the decay constants increased as the

N content in litter increased. Additionally, k tended to

increase with increasing carbohydrate content in litters and

to decrease with increasing lignin content. Berg (1986)

suggested and Taylor et al. (1989) demonstrated that the

influence of N on decomposition rates should decrease as

decay proceeds. Our results would suggest that the C/N and

lignin/N ratios determine the speed of mass loss in the first

phase of litter decomposition. On the other hand, the car-

bohydrate and lignin content of litter species seem to

influence both the early and the late stages of the decom-

position process. a2 (which, in the cases of L. styraciflua

and P. taeda is the asymptotic value of the decomposition

curves) was found to increase with the carbohydrate con-

tent and to have a negative trend with the lignin content. It

also depends on the litter N concentration, with higher a2

values when the N content is higher.

With regard to the measurements of the d13C of the

respired CO2, very interestingly, an isotopic fractionation

was found for the CO2 respired by the control soil: the

d13C(CO2) respired by the control soil moved toward more

negative values (Fig. 3, lower panel) becoming * 2% more

negative than the bulk soil. In the past, different fractiona-

tions were observed (Mary et al. 1992; Santruckova et al.

2000) during the decomposition of different substrates. The

trend we found may be explained in different ways:

1. Selective degradation of different compounds: more

recalcitrant compounds, that generally have more

negative d13C values, are respired in the second phase

of the decomposition process.

2. Microbial discrimination associated with the respira-

tion process.

In any case, this is an indication of isotopic ‘‘fractionation’’

mechanisms during the respiration process of the SOM.

Further mechanistic studies are needed to get an insight

into these processes.

In our experiment the assumption of no fractionation

during the respiration of litter compounds could not be

tested. Similarly to soil samples, a fractionation in the res-

piration process is likely to happen also for litter samples.

As suggested by past studies (Fernandez and Cadish 2003;

Fernandez et al. 2003; Schweizer et al. 1999), a fraction-

ation of only few units of % may occur. This is also

supported by no change in the d13C of litter substrates

(Table 1). According to this assumption, the difference

between the d13C of the respired CO2 and the d13C of the

litter substrate can be considered negligible in comparison

to the difference between the d13C of litter and soil samples

(Dd13C*26%). A fractionation of *2% (which is ca. 10%

of variation of d13C compared to Dd13C) would not cause a

significant variation in the cumulative C–CO2 respired by

litter samples. For this reason the influence of the frac-

tionation process on the d13C of the respired CO2 is low and

the assumption of no fractionation can be considered valid.

C lost during decomposition: C input into the soil

Berg (2000) estimated the C input from decomposing litter

to soil by modelling the limit value at which mass loss

stabilizes (or becomes extremely slow). Over the 130 cases

investigated, he found that the higher the N concentration

of fresh litter (and the lower the C/N ratio) the more

organic matter left from the decomposing litter entered the

SOM (Berg 2000).

Using isotopic methodologies, no assumption about the

asymptotic mass loss from litter is needed and the input of

organic C in soil can be measured simply by measuring

isotopic variation in soil.

Oecologia (2007) 154:155–166 163

123

In our study, the C input from decomposing litter to soil

was in the range: 10.5–18.1% of C loss from L. styraciflua,

8.2–21.8% from C. canadensis and 0–13.9% for P. taeda

(Table 3). There was no significant difference among these

values.

The C input was found to be proportional to the total C

loss from litter (Fig. 4; r = 0.780, P < 0.01).

A positive trend with the carbohydrate content and an

opposite one with the lignin content of each litter species

was found. Thus the hypothesis based on Fritsche (1998)

(the higher the lignin concentration of litter the higher the

fraction of respired CO2 over the total C loss) was not sup-

ported. Finally no correlation was found between the

asymptotic value (a2) of litter respiration rates and the C

input in soil. Again this would suggest that the assumption

from Berg (2000) of the equality between mass loss limit and

C input into the soil could be wrong and should be revised.

In summary, there was no dependence of the C lost by

the litter samples (as CO2 and as an input to the SOM) on

the chemical characteristics of the litter species. The linear

fit shown in Fig. 4 can be used to quantify the average C

input (%) over the total C lost by litters. The slope of this fit

gives an average ratio of C input (g) to total C loss (g) that

represents the percentage of C input to SOM from

decomposing litter. This is equal to 13 ± 3% (the error is

derived from the fitting procedure). Since the ‘‘environ-

mental factors’’ (such as temperature and humidity) were

constant in our systems and there was no fragmentation of

leaves due to soil macrofauna, this number gives us a

quantification of the microbial assimilation efficiency of C.

In other words, our results suggest that the microbial

assimilation efficiency of C is not dependent on the litter’s

chemical characteristics. This could be explained by

hypothesizing that the rate of decomposition of litter is

strictly dependent on the microbial community acting in

the decomposition process. If so, in our experimental sys-

tems, independently from the litter species, the same

microbial community could have been developed because

the soil used was the same. In any case, to our knowledge,

this is the best estimate for the C input from decomposing

leaf litter to the SOM and, even if it is derived from a

laboratory test in closed conditions, it could be used in

terrestrial models as a fixed percentage of the total C lost

by leaf litter. Obviously, experiments are needed to confirm

the results under field conditions and to quantify the C

input due to other process such as fragmentation and

incorporation of litter fragments in soil aggregates.

Conclusion

Our results suggest that litter with an initially fast decay

rate may have, over a long period, higher mass remaining.

That is, a litter with high percentages of labile compounds

(that would show a rapid mass loss in the early stage of the

decomposition process), may have a smaller total mass loss

(due to more recalcitrant compounds which remain). In this

respect, the definition of high-quality litter as a litter with a

fast decay rate could be under discussion, for example by

altering the definition to give one definition for the early

and one definition for the late stages of the decomposition

process.

A dependence of the fraction of litter C lost as an input

to SOM by litter on the total amount of C loss was evident.

This suggests that a fixed percentage of C over the total C

lost by the decomposing litter enters the SOM. In this

respect, the hypothesis that the quantity of litter-derived C

incorporated into SOM is not proportional to the total C

loss was not verified. A larger number of litter species

needs to be investigated to confirm these results as well as

other experiments are needed to get insight into the

decomposition process and to decrease the large uncer-

tainties of litter decomposition models in the description of

the dependence of litter quality on litter decomposition and

particularly on C transfers to SOM.

Acknowledgements We would like to thank ESF (SIBAE program)

for a grant given to Mauro Rubino to collaborate at the Max Planck

Institute for Biogeochemistry (Jena). The work was partially sup-

ported by the EU’s CARBOEUROPE IP. We would also like to thank

Federica Farina for her help in the lab and the anonymous reviewers

whose suggestions greatly improved the original manuscript. To the

best of their knowledge, the authors of the present manuscript declare

that the experiment described above complied with Italian law.

References

Anderson JM, Ineson P (1982) A soil microcosm system and its

applications to measurements of respiration and nutrient leach-

ing. Soil Biol Biochem 14:415–416

Balesdent J, Mariotti A, Guillet B (1987) Natural 13C abundance as a

trace for studies of soil organic matter dynamics. Soil Biol

Biochem 19:25–30

0.00

0.02

0.04

0.06

0.08

0.0 0.2 0.4 0.6

Total C loss (g)

)g( lios ni tupni na sa ssol C

L. styraciflua

C. canadensis

P. taeda

Fig. 4 Litter-derived C input to soil organic matter as a function of

total C lost by each litter species. The best fit equation is expressed

by: y ¼ 0:12916x þ 0:00003 (r = 0.780)

164 Oecologia (2007) 154:155–166

123

Berg B (1986) Nutrient release from litter and humus in coniferous

forest soils—a mini review. Scand J For Res 1:359–369

Berg B (2000) Litter decomposition and organic matter turnover in

northern forest soils. For Ecol Manage 133:13–22

Berg B, Agren GI (1984) Decomposition of needle litter and its

organic chemical components: theory and field experiments.

Long-term decomposition in a Scots pine forest. III. Can J Bot

62:2880–2890

Berg B, Ekhbom G (1991) Litter mass loss rates and decomposition

patterns in some needles and leaf litter types. Long-term

decomposition in a Scots pine forest. VII. Can J Bot 69:1449–

1456

Berg B, Staaf H (1980) Decomposition rates and chemical changes in

decomposing needle litter of Scots pine. II. Influence of chemical

composition. In: Persson T (ed) Structure and function of

northern coniferous forests, vol 32. Ecological Bulletins, NFR,

pp 373–390

Berg B, Staaf H, Wessen B (1987) Decomposition and nutrient

release in needle litter from nitrogen-fertilized Scots pine (Pinussylvestris) stands. Scand J For Res 2:399–415

Bertolini T, et al. (2005) Optimized sample preparation for isotopic

analyses of CO2 in air: systematic study of precision and

accuracy dependence on driving variables during CO2 purifica-

tion process. J Mass Spectrom 40:1104–1108

Bocock KL, Gilbert O (1957) The disappearance of leaf litter under

different woodland conditions. Plant Soil 9:179–185

Chapin IFS, Matson PA, Mooney HA (2002) Principles of terrestrial

ecosystem ecology. Springer, New York

Corbeels M (2001) Plant litter and decomposition: general concepts

and model approaches. NEE Workshop proceedings

Cortez J, Bouche M (2001) Decomposition of mediterranean leaf

litters by Nicodrilus meridionalis (Lumbricidae) in laboratory

and field experiments. Soil Biol Biochem 33:2023–2035

Cortez J, Demard JM, Bottner P, Jocteur Monrozier L (1996)

Decomposition of Mediterranean leaf litters: a microcosm

experiment invetigating relationships between decomposition

rates and litter quality. Soil Biol Biochem 28:443–452

Cotrufo MF, Ineson P, Rowland P (1994) Decomposition of tree leaf

litters grown under elevated CO2: effect of litter quality. Plant

Soil 103:121–130

Cotrufo MF, Drake B, Ehleringer JR (2005) Palatability trials on

hardwood leaf litter grown under elevated CO2: a stable carbon

isotope study. Soil Biol Biochem 37:1105–1112

Del Galdo I, Six J, Peressotti A, Cotrufo MF (2003) Assessing the

impact of land-usechange on soil C sequestration in agricultural

soils by means of organic matter fractionation and stable C

isotopes. Global Change Biol 9:1204–1213

Enriquez S, Duarte CM, Sand-Jensen K (1993) Patterns in decom-

position rates among photosynthetic organisms: the importance

of detritus C:N:P content. Oecologia 94:457–471

Falconer GJ, Wright JW, Beall HW (1933) The decomposition of

certain type of fresh litter under field conditions. Am J Bot

20:196–203

Fernandez I, Cadish G (2003) Discrimination against 13C during

degradation of simple and complex substrates by two white rot

fungi. Rapid Commun Mass Spectrom 17:2614–2620

Fernandez I, Mahieu N, Cadish G (2003) Carbon isotopic fraction-

ation during decomposition of plant materials of different

quality. Global Biogeochem Cycles 3:1075–1086

Fritsche W (1998) Umwelt-Mikrobiologie. Fisher, Jena

Gartner TB, Cardon ZG (2004) Decomposition dynamics in mixed-

species leaf litter. Oikos 104:230–246

Gholz HL, Wedin DA, Smitherman SM, Harmon ME, Parton MJ

(2000) Long-term dynamics of pine and hardwood litter in

contrasting environments: toward a global model of decompo-

sition. Global Change Biol 6:751–765

Gillon D, Joffre R, Ibrahima A (1999) Can litter decomposability be

predicted by near infrared reflectance spectroscopy? Ecology

80:175–186

Gleixner G, Schmidt HL (1998) On-line determination of group

specific isotope ratios in model compounds and acquatic humic

substances by coupling pyrolysis to GC-C-IRMS. In: Stankiew-

ics BA, Van Bergen PF (ed) 214th National Meeting of the

American Chemical Society. American Chemical Society, Las

Vegas, Nev., pp 34–45

Gleixner G, Bol B, Balesdent J (1999) Molecular insight into soil

carbon turnover. Rapid Commun Mass Spectrom 13:1278–1283

Gleixner G, Czimczik CJ, Kramer C, Luhker B, Schmidt WIM (2001)

Plant compounds and their turnover and stabilization as soil

organic matter. In: Schulze EDHM, Harrison S, Holland E,

Lloyd J, Prentice IC, Schimel D (eds) Global biogeocemical

cycles in the climate system. Academic Press, Jena, pp 201–215

Harris D, Horwarth WR, Van Kessel C (2001) Acid fumigation of

soils to remove carbonates prior to total organic carbon or

carbon-13 isotopic analasys. Soil Sci Soc Am J 65:1853–1856

Heal O, Anderson J, Swift M (1997) Plant litter quality and

decomposition: an historical overview. In: Cadish G, Giller K

(eds) Driven by nature: plant litter quality and decomposition.

CAB International, Wallingford, pp 3–30

Hendrey GR, Lewis KF, Nagy J (1993) Free air carbon dioxide

enrichement: development, progress, results. Vegetation 104/

105:17–31

Hungate BA, Jackson RB, Field CB, Chapin Iii FS (1996) Detecting

changes in soil carbon in CO2 enrichment experiments. Plant

Soil 187:135–145

Jenkinson DS, Adams DE, Wild A (1991) Model estimates of CO2

emissions from soil in response to global warming Nature

351:304–306

Joffre R, Agren GI, Gillon D, Bosatta E (2001) Organic matter quality

in ecological studies: theory meets experiment. Oikos 93:451–458

Leavitt SW, et al. (2001) Stable-carbon isotopes and soil organic

carbon in wheat under CO2 enrichment. New Phytol 150:305–

314

Lichter J, et al. (2005) Soil carbon sequestration and turnover in a

pine forest after six years of atmospheric CO2 enrichment.

Ecology 86:1835–1847

Liski J, Ilvesniemi H, Makela A, Starr M (1998) Model analysis of the

effects of soil age, fires and harvesting on the carbon storage of

boreal forest soils. Eur J Soil Sci 49:407–416

Mary B, Mariotti A, Morel JL (1992) Use of 13C variations at natural

abundance for studing the biodegradation of root mucilage, roots

and glucose in soil. Soil Biol Biochem 24:1065–1072

Matthews E (1997) Global litter production, pools and turnover times:

estimates from measurements data and regression models. J

Geophys Res Atmos 102:18,771–718,800

McClaugherty CA, Berg B (1987) Cellulose, lignin and nitrogen

concentrations as rate regulating factors in late stages of forest

litter decomposition. Pedobiologia 30:101–112

Melillo JM, Aber JD, Muratore JF (1982) Nitrogen and lignin control

of hardwood leaf litter decomposition dynamics. Ecology

63:621–626

Melin E (1930) Biological composition of some types of litter from

North American forests. Ecology 11:72–101

Millar HC, Smith FB, Brown PE (1948) The rate of decomposition of

various plant materials in soils. J Am Soc Agron 28:914–923

Minderman G (1968) Addition, decomposition and accumulation of

organic matter in forests. J Ecol 56:355–362

Parton WJ, Ojima DS, Cole CV, Schimel DS (1994) A general model

for soil organic matter dynamics: sensitivity to litter chemistry,

texture and management. In: Bryant RB, Arnold RW (eds)

Quantitative modeling of soil forming processes, vol 39. SSSA

special publication. SSSA, Madison, Wis., pp 147–167

Oecologia (2007) 154:155–166 165

123

Perez-Harguindeguy N, Diaz DS, Cornelissen JHC, Vendramini F,

Cabido M, Castellanos A (2000) Chemistry and toughness

predict leaf litter decomposition rates over a wide spectrum of

functional types and taxa in central Argentina. Plant Soil

218:21–30

Pope MI, Judd MJ (1977) Differential thermal analysis. Academic

Press, London

Santruckova H, Bird MI, Lloyd J (2000) Microbial processes and

carbon-isotope fractionation in tropical and temperate grassland

soils. Funct Ecol 14:108–114

Schweizer M, Fear J, Cadish G (1999) Isotopic (13C) fractionation

during plant residue decomposition and its implications for soil

organic matter studies. Rapid Commun Mass Spectrom 13:1284–

1290

Swift MJ, Heal OW, Anderson JM (1979) Decomposition in

terrestrial ecosystems. Blackwell, Oxford

Taylor BR, Parkinson D, Parsons WFJ (1989) Nitrogen and lignin

content as predictor of litter decay rates: a microcosm test.

Ecology 70:97–104

Zech W, Kogel-Knaber I (1994) Patterns and regulation of organic

matter trasformation in soils: litter decomposition and humifi-

cation. In: Schulze ED (ed) Flux control in biological system:

from enzymes to populations and ecosystems. Academic Press,

San Diego, Calif., pp 303–335

166 Oecologia (2007) 154:155–166

123