Contributions of roots, rhizodeposits, and soil organic matter to CO2
efflux from maize rhizosphere as revealed by 13C and 14C tracer
methods
Bestimmung der Beiträge der Wurzeln, Rhizodeposite und
organischen Bodensubstanz zum CO2-Efflux einer Maisrhizosphäre
durch 13C- und 14C-Markierungsmethoden
Dissertation zur Erlangung des Doktorgrades
der Naturwissenschaften (Dr. rer. nat.)
Fakultät Naturwissenschaften
Universität Hohenheim
Institut für Botanik
Institut für Bodenkunde und Standortslehre
vorgelegt von
Martin Werth
aus Haan
2007
2
Dekan bzw. Dekanin: Prof. Dr. rer. nat. Heinz Breer
1. berichtende Person: Prof. Dr. rer. nat. Yakov Kuzyakov
2. berichtende Person: Prof. Dr. rer. nat. Manfred Küppers
Eingereicht am: 01.06.2007
Mündliche Prüfung am: 06.11.2007
Die vorliegende Arbeit wurde am 25.09.2007 von der Fakultät Naturwissenschaften der Universität
Hohenheim als „Dissertation zur Erlangung des Doktorgrades der Naturwissenschaften“ angenommen.
3
Table of contents
Abbreviations .......................................................................................................................................4
1
Introduction ..................................................................................................................................5
2
Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
Plant and Soil (2006), 284: 319-333...........................................................................................17
3 Three-source partitioning of CO2 efflux from soil planted with maize by 13C natural abundance
fails due to inactive microbial biomass
Soil Biology & Biochemistry (2006), 38: 2772-2781..................................................................39
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
Soil Biology & Biochemistry (2007), in press, doi:10.1016/j.soilbio.2007.09.022.....................59
5 Partitioning of CO2 efflux from soil planted with maize by 13C natural abundance and root
exclusion
Journal of Plant Nutrition and Soil Science (2007), under review.............................................83
6
Final conclusions and perspectives ..........................................................................................107
7
Summary ..................................................................................................................................115
8
Zusammenfassung....................................................................................................................118
Acknowledgements ..........................................................................................................................121
Curriculum vitae...............................................................................................................................122
Note: Chapter 5 is written in American English; all remaining chapters are written in British English.
4
Abbreviations
ANOVA
analysis of variance
DNS
10 x diluted nutrient solution
DW
deionised water
EA
elemental analyzer
IRMS
isotope ratio mass spectrometer
LSD
least significant difference
M
molar concentration (mol l–1)
MB
microbial biomass
n.d.
not determined
n.e.
not existing
NS
nutrient solution
ns
not significant
RDR
root-derived respiration
RMR
rhizomicrobial respiration
RR
root respiration
SD
standard deviation
SOM
soil organic matter
SOMD
soil organic matter decomposition
TIC
total inorganic carbon
TOC
total organic carbon
1
Introduction
6
1 Introduction
Contributions to total CO2 efflux from soil
Soils contain the largest terrestrial carbon (C) pool on Earth amounting 1460 to 1580 Gt, i.e. on
average more than two-thirds of the total C in terrestrial ecosystems (Schimel, 1995; Grace, 2004).
Through soil respiration, they contribute an annual flux of CO2 to the atmosphere that is about 10
times the flux from fossil fuel combustion (Schlesinger, 1997). The Gross Primary Productivity of
the vegetation is about 120 Gt C yr–1, while about 60 Gt C yr–1 are released again into the atmosphere by respiration of autotrophic plants (Grace, 2004). Consequently, about 60 Gt C yr–1 enter the
soils from the atmosphere via plant residues (this is equal to Net Primary Productivity), but approximately the same amount of carbon is released again into the atmosphere by respiration from
soil heterotrophic microorganisms (‘decomposition’). Due to almost equal carbon inputs and outputs on a global scale, soils can function both as a carbon source or sink. Owing to the size of these
fluxes, even small changes in the rate of soil respiration could have large effects on the atmospheric
CO2 concentration. Several studies demonstrate that increased rates of soil respiration may result
from increases in soil temperature (Jenkinson et al., 1991; Kirschbaum, 1995; Winkler et al., 1996;
Christensen et al., 1997) and/or atmospheric CO2 (Johnson et al., 1994; Hungate et al., 1997; Ball
and Drake, 1998). In order to clarify whether a soil functions as a CO2 source or sink, it is important
to clarify which pools feed into soil respiration and to quantify these pools. Generally, the CO2 efflux from soil can be separated into five components (Kuzyakov, 2006): (1) root respiration, i.e.
respiration of assimilates by roots of autotrophic plants, (2) rhizomicrobial respiration, i.e. respiration of rhizodeposits by heterotrophic microorganisms in the rhizosphere, (3) decomposition of
dead plant residues by heterotrophic microorganisms, (4) priming effect, i.e. plant-induced additional (or reduced) decomposition of soil organic matter (SOM) by heterotrophic microorganisms,
and (5) SOM decomposition by heterotrophic microorganisms.
In the absence of plant residues and assuming a low contribution of priming effects in fertilized
agricultural soils or soils under controlled laboratory conditions (Cheng and Coleman, 1990; Paterson and Sim, 1999; 2000), the three main components of CO2 efflux will be reduced to (1) root respiration, (2) rhizomicrobial respiration and (3) SOM decomposition. The first two fluxes and at
least a part of the third flux derive from the rhizosphere, which can be best defined as the volume of
soil around living plant roots that is influenced by root activity (Darrah, 1993; Hinsinger, 1998).
‘Root respiration’, one of these rhizosphere CO2 fluxes, derives from the decomposition of plant
assimilates within the root cells. After assimilation of CO2, organic molecules in plants are built up,
especially starch and sugars, which are used as a source of energy for the plants themselves. A part
of these compounds is transported into the roots, where their energy is released by respiratory processes with the carbon being returned to the atmosphere as CO2 from root respiration.
Besides this CO2 flow from the roots, another part of the assimilated carbon gets released into
the soil by rhizodeposits, i.e. all organic substances entering the soil organic matter pool through the
living roots. Three broad types of rhizodeposits can be determined (Brady and Weil, 1999; Nguyen,
1 Introduction
7
2003). First, low-molecular-weight organic compounds are passively exuded by root cells, including organic acids, sugars, amino acids, and phenolic compounds. Second, high-molecular-weight
mucilages actively secreted by root-cap cells and epidermal cells near apical zones form a substance
called mucigel when mixed with microbial cells and clay particles. Third, cells from the root cap
and epidermis continually slough off as the root grows or get digested by bacteria. These lysates
enrich the rhizosphere with a wide variety of cell contents. Microorganisms in the rhizosphere decompose the rhizodeposits leading to ‘rhizomicrobial respiration’. The sum of both flows – root and
rhizomicrobial respiration – forms ‘root-derived CO2’. The term ‘rhizosphere respiration’ has also
often been used when considering root-derived CO2, but this term refers to the location of respiration rather than to the substrate being used (Kuzyakov and Larionova, 2005).
Soil organic matter, which is built up by the residues of dead plants, rhizodeposits, and dead
animals gets partly decomposed by soil microorganisms leading to the third contribution to total
CO2 efflux from soil: ‘SOM-derived CO2’. This CO2 source can be regarded as soil-derived CO2
flux, while the two first-mentioned sources can be described as plant-derived CO2 fluxes. While an
increase in the SOM-derived CO2 flux has an impact on climate change, the plant-derived CO2 flux
cannot be considered as additional greenhouse gas.
Partitioning methods for CO2 efflux from soil
Carbon dioxide derived from SOM decomposition and that derived from the roots can be partitioned and quantified by isotopic labelling of plants with 14C or 13C isotopes and tracing the label in
root-derived CO2 (Warembourg and Paul, 1977; Andrews et al., 1999). The difference between the
labelled fraction and the total CO2 efflux represents CO2 from SOM decomposition. Besides artificial labelling techniques, the difference in the natural abundances of 13C in C3 and C4 plants can
also be used as natural carbon tracer to calculate root- and SOM-derived CO2 contributions by mass
balance equations (Cheng, 1996; Qian et al., 1997; Rochette and Flanagan, 1997; Ekblad and Högberg, 2001; Kuzyakov and Cheng, 2001; Midwood et al., 2006). Non-isotopic methods to separate
root- from SOM-derived CO2, such as component integration (Larionova et al., 2003), excised roots
(Reich et al., 1998; Burton and Pregitzer, 2002), root exclusion (Rochette et al., 1999; Larionova et
al., 2006), shading or clipping (Craine et al., 1999; Paterson and Sim, 1999), tree girdling (Högberg
et al., 2001), or trenching (Kelting et al., 1998), have also been used. The results vary strongly depending on plants, soils, and environmental and experimental conditions. For example, Warembourg and Paul (1977) found low contributions (19%) of root-derived CO2 to the total CO2 efflux
from soil by in situ 14C labelling of Canadian prairie grass. In contrast, under controlled conditions,
Chen et al. (2006) reported very high contributions of root-derived CO2 to the total CO2 efflux, with
values of up to 99% in a ryegrass (Lolium perenne L.) rhizosphere. Various studies with grass species have found results within this range with an average root-derived CO2 contribution of 70 ±
27% in the laboratory and 36 ± 15% in the field (compare Table 5-1).
8
1 Introduction
It is very difficult to further differentiate between the CO2 directly derived from root respiration
and that derived from mineralisation of rhizodeposits (Killham and Yeomans, 2001). This separation of root and rhizomicrobial respiration is a major challenge in quantifying rhizosphere carbon
flows. Separation is important to quantify carbon sources for SOM and for rhizosphere microorganisms, to identify respiration of autotrophic and heterotrophic organisms, and to calculate carbon
turnover by rhizosphere microorganisms (Kuzyakov, 2004). Separation results of several studies on
various grass species reveal that root-derived CO2 consists of about 50 ± 10% for both root and rhizomicrobial respiration (compare Table 5-1).
Some attempts to separate root and rhizomicrobial respiration were tested more or less successfully (reviewed by Kuzyakov and Larionova, 2005). Most of those studies were conducted under
controlled conditions. The methods were based on pulse labelling of plants in a 14CO2 atmosphere
and tracing the 14CO2 dynamics (Kuzyakov et al., 1999; Kuzyakov et al., 2001; Kuzyakov and Domanski, 2002), isotopic dilution (Cheng et al., 1993) or various treatments with 14C labelled plants
and rhizodeposits (Johansson, 1992; Swinnen, 1994). The only field studies that attempted to separate root and rhizomicrobial respiration were based on trenching and excised roots (Kelting et al.,
1998), on shading and clipping and excised roots (Craine et al., 1999), or on root exclusion and
component integration (Larionova et al., 2006). All of these approaches very strongly disturbed the
soil and/or the roots, making the relevance of these results questionable.
In this thesis three methods were used to separate root- and SOM-derived carbon in the CO2 efflux from soil and in the microbial biomass: (1) root exclusion, (2) artificial 14C labelling, and (3)
natural 13C labelling. In order to determine plant-derived carbon contributions with the root exclusion approach, the total CO2 efflux from bare fallow soil was subtracted from the efflux from soil
planted with maize. The two isotopic approaches will be considered in more detail in the following
two sections.
14
C labelling
The radioactive 14C isotope is continuously produced in the upper atmosphere by neutrons from the
cosmic radiation hitting on nitrogen leading to the release of a proton. The 14C isotope decays by
emitting a negatively charged beta particle (β–) and an antineutrino and results in elemental nitrogen. In this study, the β–-emission is measured using liquid scintillation counting. Liquid samples
are obtained by trapping CO2 in NaOH solution, collecting exudates in nutrient solution, extracting
microbial biomass with K2SO4 solution, or trapping CO2 from plant and soil samples in an organic
absorbens after combustion in an oxidizer unit. An aliquot of the sample solution is mixed with a
scintillation cocktail containing solvents and photosensitive chemicals (fluors). The β–-particles
emitted from the sample activate the fluors, creating flashes of light, which are detected by paired
photomultiplier tubes, positioned 180° apart, in the liquid scintillation counter. Quenching may interfere the measurement leading to an underestimation of decay. Therefore, the counting efficiency
has to be taken into account for converting counts per minute into disintegrations per minute, the
1 Introduction
9
actually occurring decay in the sample. Most instruments have a γ-emitting external standard built
into the counter enabling them to calculate counting efficiencies automatically. Chemiluminescence
and phospholuminescence may also interfere the measurement, but are generally prevented when
storing the samples in the dark.
For studies on carbon flows in the rhizosphere, containers with plants are sealed between shoots
and roots and plant shoots are labelled in a 14CO2-atmosphere. Either pulse or continuous labelling
can be applied depending on the objectives and facilities. Pulse labelling, compared with continuous
labelling, has the advantage of being easier to handle (Whipps, 1990), providing more information
on the recent photosynthate distribution at specific developmental stages of the plants (Meharg and
Killham, 1990; Swinnen et al., 1994a), and being suitable for kinetic investigations of 14CO2 evolution from the soil (Swinnen et al., 1994a; Kuzyakov et al., 1999; Domanski et al., 2001). Carbon
dioxide from soil respiration is being trapped following the labelling pulse. After a distinct time, 14C
allocation to shoots, roots, exudates, soil organic matter, microbial biomass, CO2 etc. can be determined by destructive sampling of the containers and further analytical preparations of these fractions. Since partitioning patterns change during plant growth, the 14C distribution at one stage of
development cannot be applied to another, or to a whole growth period. However, a series of labelling pulses applied at regular intervals during plant growth provides a reasonable estimate of the
cumulative below-ground C input (Gregory and Atwell, 1991; Swinnen et al., 1994b).
Continuous labelling is particularly appropriate for the estimation of the total C amount transferred by the plant into below-ground soil pools during the labelling period (Meharg, 1994), e.g.
during a whole growth period. However, continuous labelling requires special equipment for exposing plants to 14CO2 with a constant specific activity over a long time period. Furthermore, the air
temperature and air humidity must be controlled inside the labelling chamber. These facilities are
expensive and limited to only a few places in the world. An alternative to artificial continuous labelling is the natural 13C labelling technique.
Natural 13C labelling
The natural abundance of the 12C and 13C stable carbon isotopes between plant species with different photosynthetic pathways can be utilised to label C flows in the rhizosphere. Approximately
98.89% of all C in nature is 12C, and 1.11% is 13C (Boutton, 1991b).
After combustion of a solid sample in an elemental analyser (EA), the isotopic composition of
the produced CO2 is determined by simultaneous collection of masses 44 (12C16O16O), 45
(13C16O16O), and 46 (12C18O16O) in an isotope ratio mass spectrometer (IRMS). Since the natural
absolute variation in the 13C/12C ratio is small, stable C isotope ratios are expressed relative to the
international PDB limestone standard as δ13C:
δ 13 C [‰] =
Rsample − RPDB
RPDB
⋅1000‰
(1-1),
10
1 Introduction
where R is the isotopic ratio of 13C/12C. RPDB is the defined isotope ratio of the limestone fossil Belemnitella americana from the Cretaceous PeeDee Formation in South Carolina, which is set to δ13C
= 0‰ as basis of the scale. It has an absolute 13C/12C ratio of 0.0112372 (Craig, 1953). A negative
δ13C value indicates that the sample is “lighter”, i.e. contains less 13C than the standard.
The 13C/12C ratio of organic carbon in terrestrial ecosystems is influenced by the C isotope fractionation occurring during photosynthesis (Wolf et al., 1994). Plants with the C3 photosynthetic
(Calvin-Benson) pathway reduce CO2 to phosphoglycerate, a 3-C compound, via the enzyme ribulose 1,5 bisphosphate carboxylase/oxygenase (Rubisco). This enzyme discriminates against 13CO2,
resulting in plant δ13C values between –22 and –32‰ and an average of –27‰ (Boutton, 1996).
The discrimination Δ in C3 photosynthesis compared to atmospheric CO2 can be explained by the
following equation (Farquhar et al., 1982):
Δ = a + (b − a) ⋅
pi
pa
(1-2),
where a is the fractionation occurring due to diffusion in air, b is the net fractionation caused by
carboxylation, and pa and pi are the ambient and intercellular partial pressures of CO2, respectively.
The enzyme phosphoenol pyruvate (PEP) carboxylase of C4 plants reduces CO2 to aspartic or
malic acid (Hatch-Slack pathway), both 4-C compounds, with lower discrimination against 13C than
Rubisco. Thus, δ13C values of C4 plants range from –9 to –17‰ with an average of –13‰ (Boutton,
1996). The discrimination in C4 photosynthesis (Farquhar, 1983) is:
Δ = a + (b4 + b3φ − a ) ⋅
pi
pa
(1-3),
where b4 is the effective discrimination by PEP carboxylase, b3 is the discrimination by Rubisco,
and φ is the proportion of the carbon fixed by PEP carboxylation that subsequently leaks out of the
bundle sheath.
Some plants with the Crassulacean acid metabolism (CAM) are able to switch between those two
photosynthetic pathways and consequently their δ13C values range from –10 to –28‰ (Boutton,
1996). C3 plant species dominate most temperate zone and all forest communities. C4 plant species
as well as CAM plants are more common in warm, arid, or semiarid environments (Boutton, 1991a;
Ehleringer, 1991). Maize (Zea mays L.) – the model plant of this thesis – is a typical widespread C4
crop species.
Once C3 or C4 plants have assimilated carbon with their typical discrimination process against
C, there exist further fractionation processes. Within plants, the δ13C values of different compounds vary. It has been observed that lignin and lipids are usually 13C-depleted compared to the
bulk plant material while sugars, amino acids and hemicelluloses are 13C-enriched (Boutton, 1996;
Hobbie and Werner, 2004). The specific enrichment of 13C in transport compounds like sucrose
leads to an enrichment of 13C in the roots (Hobbie and Werner, 2004).
13
1 Introduction
11
The isotopic composition of soil organic matter largely reflects the photosynthetic pathway type
of the vegetation growing on a certain soil. Changes from an initial C3 vegetation to a C4 vegetation
or vice versa can hence be used as ‘natural 13C labelling technique’ (Balesdent and Mariotti, 1996).
In this technique, the unique isotopic composition of the new source vegetation acts as a carbon
tracer when introduced to a soil organic matter pool or below-ground CO2 pool derived from a different source. Kuzyakov (2004; 2005) suggested a new approach using the natural 13C labelling
technique after C3–C4 vegetation change in order to separate root respiration, rhizomicrobial respiration, and soil organic matter decomposition. In this approach, the δ13C values of SOM, maize
roots, microbial biomass, and total CO2 efflux from the soil are used to determine the three fractions
of CO2, which can be calculated according to the isotopic mass balance of microbial biomass and
CO2. One main objective of this thesis was to test this approach. If present, 13C fractionation processes during root or microbial respiration have to be considered in this approach.
Exact 13C fractionations have to be determined in a single chemical reaction (see Eq. (1-2) and
(1-3)) considering the δ13C values of the educts and products (see Hobbie and Werner, 2004). In
root respiration, for example, this consideration would include δ13C values of the sugars involved in
respiration and of the emitted CO2. Instead of using single compounds, in our studies it was sufficient to consider δ13C values of bulk roots, SOM, or microbial biomass, since these were used for
CO2 partitioning. Differences in δ13C values between bulk materials and the emitted CO2 can be
explained by various transformation and decomposition processes involved with their unique isotopic fractionations caused by biologically preferred utilization of 13C-enriched (or -depleted) compounds and chemically faster or more slowly reacting isotopes. These fractionations (Δ) between a
source and a product (Lajtha and Michener, 1994) are expressed as:
Δ=
δa −δb
1 + δ b / 1000
(1-4),
where δa is the δ13C value of the source and δb is the δ13C value of the product. Since the denominator is close to 1, the simplified equation
Δ = δa −δb
(1-5),
was used. If not shown by a plus or minus sign, usually the absolute value of the fractionation was
used in this thesis.
Klump et al. (2005) reported 13C-depleted CO2 from root respiration compared to the root biomass, but this fractionation was not significant. In contrast, equal δ13C values of roots and rootderived respiration have been used in most rhizosphere CO2 studies to date (Cerling et al., 1991;
Cheng, 1996; Lin and Ehleringer, 1997; Amundson et al., 1998; Ekblad and Högberg, 2000; Fu and
Cheng, 2002). Hence, further research is needed to clarify whether 13C fractionation during root
respiration occurs or not.
In several studies it was observed, that the δ13C value of CO2 evolved during the mineralisation
of organic substrates differed significantly from the δ13C value of the substrate (Mary et al., 1992;
12
1 Introduction
Schweizer et al., 1999; Šantrůčková et al., 2000; Fernandez et al., 2003; Kristiansen et al., 2004).
However, there are also studies in which this isotopic fractionation did not occur or was considered
to be negligible (Cheng, 1996; Ekblad and Högberg, 2000; Nyberg et al., 2000; Ekblad et al.,
2002). Hence, it is still uncertain which factors control the magnitude of isotopic 13C fractionation.
According to Fernandez and Cadisch (2003), carbon isotope discrimination by heterotrophic microorganisms seems to depend on temperature, molecule isotopic distribution, chemical nature of the
substrate, metabolic pathways of carbon, and physiological conditions of microbial growth.
Objectives and outline
In this thesis, several experiments with maize planted on C3 soil or grown in nutrient solution were
conducted in the laboratory or in the field. The main objective of this thesis was to test the new CO2
partitioning approach of Kuzyakov (2004; 2005) under controlled and field conditions. During this
verification of the approach the following main methodologies and objectives were investigated:
-
To identify 13C fractionations during root respiration and their dependence on assimilate
partitioning and nutrient supply (Chapter 2).
-
To determine 13C fractionations between feeding substrate, microbial biomass, and microbially respired CO2 (Chapters 3, 4, and 5).
-
To verify the natural 13C labelling approach of Kuzyakov (2004; 2005) on partitioning of
CO2 efflux from soil into root respiration, rhizomicrobial respiration, and decomposition of
soil organic matter under controlled conditions (Chapter 3).
-
To identify the impact of 13C fractionation on CO2 partitioning results by sensitivity analysis
(Chapter 3).
-
To determine the proportions of active and passive microbial biomass and their influence on
CO2 partitioning results (Chapters 3 and 5).
-
To compare the natural 13C labelling approach and the 14C pulse labelling method in order to
determine root-derived carbon contributions to CO2 efflux and soil microbial biomass
(Chapter 4).
-
To examine the natural 13C labelling approach of Kuzyakov (2004; 2005) on partitioning of
CO2 efflux from soil into root respiration, rhizomicrobial respiration, and decomposition of
soil organic matter under field conditions (Chapter 5).
-
To compare the natural 13C labelling approach and the root exclusion method in order to determine root-derived carbon contributions to the CO2 efflux from a maize field (Chapter 5).
References
Amundson, R., Stern, L., Baisden, T., Wang, Y., 1998. The isotopic composition of soil and soilrespired CO2. Geoderma 82, 83-114.
1 Introduction
13
Andrews, J.A., Harrison, K.G., Matamala, R., Schlesinger, W.H., 1999. Separation of root respiration from total soil respiration using carbon-13 labeling during free-air carbon dioxide
enrichment (FACE). Soil Science Society of America Journal 63, 1429-1435.
Balesdent, J., Mariotti, A., 1996. Measurement of soil organic matter turnover using 13C natural
abundance. In: Boutton, T.W., Yamasaki, S.I. (Eds.), Mass Spectrometry of Soils, Marcel
Dekker, New York, USA, pp. 83-111.
Ball, A.S., Drake, B.G., 1998. Stimulation of soil respiration by carbon dioxide enrichment of
marsh vegetation. Soil Biology & Biochemistry 30, 1203-1205.
Boutton, T.W., 1991a. Stable carbon isotope ratios of natural materials: II. atmospheric, terrestrial,
marine, and freshwater environments. In: Coleman, D.C., Fry, B. (Eds.), Carbon Isotope
Techniques. Isotopic Techniques in Plant, Soil, and Aquatic Biology, Academic Press, Inc.,
San Diego, pp. 173-185.
Boutton, T.W., 1991b. Stable carbon isotope ratios of natural materials: I. sample preparation and
mass spectrometric analysis. In: Coleman, D.C., Fry, B. (Eds.), Carbon Isotope Techniques.
Isotopic Techniques in Plant, Soil, and Aquatic Biology, Academic Press, Inc., San Diego,
pp. 155-171.
Boutton, T.W., 1996. Stable carbon isotope ratios of soil organic matter and their use as indicators
of vegetation and climate change. In: Boutton, T.W., Yamasaki, S.I. (Eds.), Mass
Spectrometry of Soils, Marcel Dekker, New York, pp. 47-82.
Brady, N.C., Weil, R.R., 1999. The Nature and Properties of Soils. Prentice-Hall, Inc., Upper Saddle River, 881 pp.
Burton, A.J., Pregitzer, K.S., 2002. Measurement carbon dioxide concentration does not affect
root respiration of nine tree species in the field. Tree Physiology 22, 67-72.
Cerling, T.E., Solomon, D.K., Quade, J., Bowman, J.R., 1991. On the isotopic composition of
carbon in soil carbon dioxide. Geochimica et Cosmochimica Acta 55, 3404-3405.
Chen, C.R., Condron, L.M., Xu, Z.H., Davis, M.R., Sherlock, R.R., 2006. Root, rhizosphere and
root-free respiration in soils under grassland and forest plants. European Journal of Soil Science 57, 58-66.
Cheng, W., 1996. Measurement of rhizosphere respiration and organic matter decomposition using
natural 13C. Plant and Soil 183, 263-268.
Cheng, W., Coleman, D.C., 1990. Effect of living roots on soil organic matter decomposition. Soil
Biology & Biochemistry 22, 781-787.
Cheng, W., Coleman, D.C., Carroll, C.R., Hoffman, C.A., 1993. In situ measurement of root
respiration and soluble C concentrations in the rhizosphere. Soil Biology & Biochemistry 25,
1189-1196.
Christensen, T.R., Michelsen, A., Jonasson, S., Schmidt, I.K., 1997. Carbon dioxide and methane exchange of a subarctic heath in response to climate change related environmental manipulations. Oikos 79, 34-44.
Craig, H., 1953. The geochemistry of the stable carbon isotopes. Geochimica et Cosmochimica
Acta 3, 53-92.
Craine, J.M., Wedin, D.A., Chapin, F.S.I., 1999. Predominance of ecophysiological controls on
soil CO2 flux in a Minnesota grassland. Plant and Soil 207, 77-86.
Darrah, P.R., 1993. The rhizosphere and plant nutrition: a quantitative approach. Plant and Soil
155/156, 1-20.
Domanski, G., Kuzyakov, Y., Siniakina, S.V., Stahr, K., 2001. Carbon flows in the rhizosphere
of ryegrass (Lolium perenne). Journal of Plant Nutrition and Soil Science 164, 381-387.
14
1 Introduction
Ehleringer, J.R., 1991. 13C/12C fractionation and its utility in terrestrial plant studies. In: Coleman,
D.C., Fry, B. (Eds.), Carbon Isotope Techniques. Isotopic techniques in plant, soil, and
aquatic biology, Academic Press, Inc., San Diego, pp. 187-200.
Ekblad, A., Högberg, P., 2000. Analysis of δ13C of CO2 distinguishes between microbial respiration of added C4-sucrose and other soil respiration in a C3-ecosystem. Plant and Soil 219,
197-209.
Ekblad, A., Högberg, P., 2001. Natural abundance of 13C in CO2 respired from forest soils reveals
speed of link between tree photosynthesis and root respiration. Oecologia 127, 305-308.
Ekblad, A., Nyberg, G., Högberg, P., 2002. 13C-discrimination during microbial respiration of
added C3-, C4- and 13C-labelled sugars to a C3-forest soil. Oecologia 131, 245-249.
Farquhar, G., O'Leary, M., Berry, J., 1982. On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Functional Plant Biology
9, 121-137.
Farquhar, G.D., 1983. On the nature of carbon isotope discrimination in C4 species. Australian
Journal of Plant Physiology 10, 205-226.
Fernandez, I., Cadisch, G., 2003. Discrimination against C-13 during degradation of simple and
complex substrates by two white rot fungi. Rapid Communications in Mass Spectrometry 17,
2614-2620.
Fernandez, I., Mahieu, N., Cadisch, G., 2003. Carbon isotope fractionation during decomposition
of plant materials of different quality. Global Biogeochemical Cycles 17, 1075 ff.
Fu, S., Cheng, W., 2002. Rhizosphere priming effects on the decomposition of soil organic matter
in C4 and C3 grassland soils. Plant and Soil 238, 289-294.
Grace, J., 2004. Understanding and managing the global carbon cycle. Journal of Ecology 92, 189202.
Gregory, P.J., Atwell, B.J., 1991. The fate of carbon in pulse-labelled crops of barley and wheat.
Plant and Soil 136, 205-213.
Hinsinger, P., 1998. How do plant roots acquire mineral nutrients? Chemical processes involved in
the rhizosphere. Advances in Agronomy 64, 225-265.
Hobbie, E.A., Werner, R.A., 2004. Intramolecular, compound-specific, and bulk carbon isotope
patterns in C3 and C4 plants: a review and synthesis. New Phytologist 161, 371-385.
Högberg, P., Nordgren, A., Buchmann, N., Taylor, A.F.S., Ekblad, A., Högberg, M.N., Nyberg, G., Ottosson-Löfvenius, M., Read, D.J., 2001. Large-scale forest girdling shows that
current photosynthesis drives soil respiration. Nature 411, 789-792.
Hungate, B.A., Holland, E.A., Jackson, R.D., Chapin, F.S., Mooney, H.A., Field, C.B., 1997.
The fate of carbon in grasslands under carbon dioxide enrichment. Nature 388, 576-579.
Jenkinson, D.S., Adams, D.E., Wild, A., 1991. Model estimates of CO2 emissions from soil in
response to global warming. Nature 351, 304-306.
Johansson, G., 1992. Release of organic C from growing roots of meadow fescue (Festuca pratensis L.). Soil Biology & Biochemistry 24, 427-433.
Johnson, D., Geisinger, D., Walker, R., Newman, J., Vose, J., Elliot, K., Ball, T., 1994. Soil
pCO2, soil respiration, and root activity in CO2-fumigated and nitrogen-fertilized ponderosa
pine. Plant and Soil 165, 129-138.
Kelting, D.L., Burger, J.A., Edwards, G.S., 1998. Estimating root respiration, microbial respiration in the rhizosphere, and root-free soil respiration in forest soils. Soil Biology & Biochemistry 30, 961-968.
Killham, K., Yeomans, C., 2001. Rhizosphere carbon flow measurement and implications: from
isotopes to reporter genes. Plant and Soil 232, 91-96.
1 Introduction
15
Kirschbaum, M.U.F., 1995. The temperature dependence of soil organic matter decomposition,
and the effect of global warming on soil organic C storage. Soil Biology & Biochemistry 27,
753-760.
Klumpp, K., Schäufele, R., Lötscher, M., A., L.F., Feneis, W., Schnyder, H., 2005. C-isotope
composition of CO2 respired by shoots and roots: fractionation during dark respiration?
Plant, Cell and Environment 28, 241-250.
Kristiansen, S.M., Brandt, M., Hansen, E.M., Magid, J., Christensen, B.T., 2004. 13C signature
of CO2 evolved from incubated maize residues and maize-derived sheep faeces. Soil Biology
& Biochemistry 36, 99-105.
Kuzyakov, Y., 2004. Separation of root and rhizomicrobial respiration by natural 13C abundance:
theoretical approach, advantages, and difficulties. Eurasian Soil Science 37, S79-S84.
Kuzyakov, Y., 2005. Theoretical background for partitioning of root and rhizomicrobial respiration
by δ13C of microbial biomass. European Journal of Soil Biology 41, 1-9.
Kuzyakov, Y., 2006. Sources of CO2 efflux from soil and review of partitioning methods. Soil Biology & Biochemistry 38, 425-448.
Kuzyakov, Y., Cheng, W., 2001. Photosynthesis controls of rhizosphere respiration and organic
matter decomposition. Soil Biology & Biochemistry 33, 1915-1925.
Kuzyakov, Y., Domanski, G., 2002. Model for rhizodeposition and CO2 efflux from planted soil
and its validation by 14C pulse labelling of ryegrass. Plant and Soil 239, 87-102.
Kuzyakov, Y., Larionova, A.A., 2005. Root and rhizomicrobial respiration: A review of approaches to estimate respiration by autotrophic and heterotrophic organisms in soil. Journal
of Plant Nutrition and Soil Science 168, 503-520.
Kuzyakov, Y., Kretzschmar, A., Stahr, K., 1999. Contribution of Lolium perenne rhizodeposition
to carbon turnover of pasture soil. Plant and Soil 213, 127-136.
Kuzyakov, Y., Ehrensberger, H., Stahr, K., 2001. Carbon partitioning and below-ground translocation by Lolium perenne. Soil Biology & Biochemistry 33, 61-74.
Lajtha, K., Michener, R.H., 1994. Stable Isotopes in Ecology and Environmental Science. Blackwell Scientific Publications, Oxford, 316 pp.
Larionova, A.A., Sapronov, D., Lopes de Gerenju, V.O., Kuznetsova, L.G., Kudeyarov, V.N.,
2006. Contribution of plant root respiration to the CO2 emission from soil. Eurasian Soil
Science 39, 1127-1135.
Larionova, A.A., Yevdokimov, I.V., Kurganova, I.N., Sapronov, D.V., Kuznetsova, L.G.,
Lopes de Gerenju, V.O., 2003. Root respiration and its contribution to the CO2 emission
from soil. Eurasian Soil Science 36, 173-184.
Lin, G., Ehleringer, J.R., 1997. Carbon isotope fractionation does not occur during dark respiration in C3 and C4 plants. Plant Physiology 114, 391-394.
Mary, B., Mariotti, A., Morel, J.L., 1992. Use of 13C variations at natural abundance for studying
the biodegradation of root mucilage, roots and glucose in soil. Soil Biology & Biochemistry
24, 1065-1072.
Meharg, A.A., 1994. A critical review of labelling techniques used to quantify rhizosphere carbonflow. Plant and Soil 166, 55-62.
Meharg, A.A., Killham, K., 1990. Carbon distribution within the plant and rhizosphere in laboratory and field-grown Lolium perenne at different stages of development. Soil Biology & Biochemistry 22, 471-477.
Midwood, A.J., Gebbing, T., Wendler, R., Sommerkorn, M., Hunt, J.E., Millard, P., 2006.
Collection and storage of CO2 for 13C analysis: an application to separate soil CO2 efflux
into root- and soil-derived components. Rapid Communications in Mass Spectrometry 20,
3379-3384.
16
1 Introduction
Nguyen, C., 2003. Rhizodeposition of organic C by plants: mechanisms and controls. Agronomie
23, 375-396.
Nyberg, G., Ekblad, A., Buresh, R.J., Högberg, P., 2000. Respiration from C3 plant green manure added to a C4 plant carbon dominated soil. Plant and Soil 218, 83-89.
Paterson, E., Sim, A., 1999. Rhizodeposition and C-partitioning of Lolium perenne in axenic culture affected by nitrogen supply and defoliation. Plant and Soil 216, 155-164.
Paterson, E., Sim, A., 2000. Effect of nitrogen supply and defoliation on loss of organic compounds from roots of Festuca rubra. Journal of Experimental Botany 51, 1449-1457.
Qian, J.H., Doran, J.W., Walters, D.T., 1997. Maize plant contributions to root zone available
carbon and microbial transformations of nitrogen. Soil Biology & Biochemistry 29, 14511462.
Reich, P.B., Walters, M.B., Tjoelker, M.G., Vanderklein, D., Buschena, C., 1998. Photosynthesis and respiration rates depend on leaf and root morphology and nitrogen concentration in
nine boreal tree species differing in relative growth rate. Functional Ecology 12, 395-405.
Rochette, P., Flanagan, L.B., 1997. Quantifying rhizosphere respiration in a corn crop under field
conditions. Soil Science Society of America Journal 61, 466-474.
Rochette, P., Flanagan, L.B., Gregorich, E.G., 1999. Separating soil respiration into plant and
soil components using analyses of the natural abundance of carbon-13. Soil Science Society
of America Journal 63, 1207-1213.
Šantrůčková, H., Bird, M.I., Frouz, J., Šustr, V., Tajovský, K., 2000. Natural abundance of 13C
in leaf litter as related to feeding activity of soil invertebrates and microbial mineralisation.
Soil Biology & Biochemistry 32, 1793-1797.
Schimel, D.S., 1995. Terrestrial ecosystems and the carbon cycle. Global Change Biology 1, 77-91.
Schlesinger, W.H., 1997. Biogeochemistry: An analysis of global change. Academic Press, San
Diego, 588 pp.
Schweizer, M., Fear, J., Cadisch, G., 1999. Isotopic (13C) fractionation during plant residue decomposition and its implications for soil organic matter studies. Rapid Communications in
Mass Spectrometry 13, 1284-1290.
Swinnen, J., 1994. Evaluation of the use of a model rhizodeposition technique to separate root and
microbial respiration in soil. Plant and Soil 165, 89-101.
Swinnen, J., van Veen, J.A., Merckx, R., 1994a. 14C pulse-labelling of field-grown spring wheat:
an evaluation of its use in rhizosphere carbon budget estimations. Soil Biology & Biochemistry 26, 161-170.
Swinnen, J., van Veen, J.A., Merckx, R., 1994b. Rhizosphere carbon fluxes in field-grown spring
wheat: model calculations based on 14C partitioning after pulse-labeling. Soil Biology & Biochemistry 26, 171-182.
Warembourg, F.R., Paul, E.A., 1977. Seasonal transfers of assimilated 14C in grassland: plant
production and turnover, soil and plant respiration. Soil Biology & Biochemistry 9, 295-301.
Whipps, J.M., 1990. Carbon economy. In: Lynch, J.M. (Ed.) The Rhizosphere, John Wiley and
Sons Ltd., Chichester, pp. 59-97.
Winkler, J.P., Cherry, R.S., Schlesinger, W.H., 1996. The Q10 relationship of microbial respiration in a temperate forest soil. Soil Biology & Biochemistry 28, 1067-1072.
Wolf, D.C., Legg, J.O., Boutton, T.W., 1994. Isotopic methods for the study of soil organic matter
dynamics. In: Weaver, R.W., Angle, S., Bottomley, P., Bezdicek, D., Smith, S., Tabatabai,
A., Wollum, A. (Eds.), Methods of Soil Analysis, Part 2, Microbiological and Biochemical
Properties. Soil Science Society of America Book Series, No. 5, Soil Sci. Soc. Am., Inc.,
Madison, pp. 865-906.
2
Assimilate partitioning affects
assimilated carbon in maize
13
C fractionation of recently
Martin Werth and Yakov Kuzyakov
Plant and Soil (2006), 284: 319-333
includes four tables and three figures
with kind permission of Springer Science and Business Media
18
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
Abstract
Coupling 13C natural abundance and 14C pulse labelling enabled us to investigate the dependence of
13
C fractionation on assimilate partitioning between shoots, roots, exudates, and CO2 respired by
maize roots. The amount of recently assimilated C in these four pools was controlled by three levels
of nutrient supply: full nutrient supply (NS), ten times diluted nutrient supply (DNS), and deionised
water (DW). After pulse labelling of maize shoots in a 14CO2 atmosphere, 14C was traced to determine the amounts of recently assimilated C in the four pools and the δ13C values of the four pools
were measured. Increasing amounts of recently assimilated C in the roots (from 8 to 10% of recovered 14C in NS and DNS treatments) led to a 0.3‰ 13C enrichment from NS to DNS treatments. A
further increase of C allocation in the roots (from 10 to 13% of recovered 14C in DNS and DW
treatments) resulted in an additional enrichment of the roots from DNS to DW treatments by 0.3‰.
These findings support the hypothesis that 13C enrichment in a pool increases with an increasing
amount of C transferred into that pool. δ13C of CO2 evolved by root respiration was similar to that
of the roots in DNS and DW treatments. However, if the amount of recently assimilated C in root
respiration was reduced (NS treatment), the respired CO2 became 0.7‰ 13C-depleted compared to
roots. Increasing amounts of recently assimilated C in the CO2 from NS via DNS to DW treatments
resulted in a 1.6‰ δ13C increase of root respired CO2 from NS to DW treatments. Thus, for both
pools, i.e. roots and root respiration, increasing amounts of recently assimilated C in the pool led to
a δ13C increase. In DW and DNS plants there was no 13C fractionation between roots and exudates.
However, high nutrient supply decreased the amount of recently assimilated C in exudates compared to the other two treatments and led to a 5.3‰ 13C enrichment in exudates compared to roots.
We conclude that 13C discrimination between plant pools and within processes such as exudation
and root respiration is not constant but strongly depends on the amount of C in the respective pool
and on partitioning of recently assimilated C between plant pools.
Abbreviations:
NS – nutrient solution; DNS – 10 x diluted nutrient solution; DW – deionised
water
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
19
Introduction
Labelling with 13C or 14C isotopes can be used to balance assimilated carbon (C) in a plant–soil system. These techniques are especially important because below-ground C fluxes and turnover of
root-derived C in soil cannot be fully quantified without tracers. C translocation by plants into the
soil and C partitioning of rhizodeposits and rhizosphere respiration can be observed by these tracers
and separated from C of native soil organic matter. Total rhizosphere respiration has been quantified by either continuous 14C labelling (Barber and Martin, 1976; Whipps and Lynch, 1983; Liljeroth et al., 1994) or 14C pulse labelling (Cheng et al., 1993; Kuzyakov et al., 1999; Kuzyakov,
2002). Advantages and disadvantages of the two methods have been discussed in detail by Kuzyakov and Domanski (2000) and Kuzyakov (2001). C fluxes from the plant to the soil or to any other
growth medium (such as a nutrient solution) and the CO2 efflux from the soil can be traced after
every single pulse of a repeated 14C pulse labelling. Compared to continuous labelling, a series of
labelling pulses produces more information about C translocations in the plant–soil system
(Warembourg and Estelrich, 2000).
Artificial labelling entails many methodological difficulties and is mainly limited to laboratory
studies. On-site 14C labelling often requires obtaining special permissions, but has been performed
in some pulse labelling studies (Swinnen et al., 1994a; b). 13C labelling in the field has been done
by pulse labelling (Stewart and Metherell, 1999) or by continuous labelling, e.g. in free air carbon
dioxide enrichment (FACE) studies (Søe et al., 2004). As a reasonable alternative to artificial labelling, natural 13C labelling by planting C4 plants on a soil developed under C3 vegetation or vice
versa has been frequently used in the last 15 years (Balesdent et al., 1987; Cheng, 1996; Rochette
and Flanagan, 1997; Rochette et al., 1999; Gerzabek et al., 2001; Cheng et al., 2003; John et al.,
2003). It has been used to estimate below-ground C input, its partitioning and separation of CO2
sources from soil. However, the results of C balance and partitioning studies obtained by natural 13C
labelling and FACE studies can be biased by isotopic fractionation involving for example transport
of assimilates, rhizodeposition, and root respiration. In studies with C3 leaves, there is a strong evidence that dark respiratory CO2 was significantly enriched in 13C compared to the putative substrate
(Duranceau et al., 1999; Ghashghaie et al., 2001; Ghashghaie et al., 2003; Tcherkez et al., 2003).
Other studies, however, have shown no difference in δ13C values of total root mass and of root respiration (Cerling et al., 1991; Amundson et al., 1998) or of total rhizosphere respiration (Cheng,
1996; Fu and Cheng, 2002). Incubation experiments reveal that CO2 respired by microbial decomposition of root residues was depleted in 13C by 1 to 10‰ (Mary et al., 1992; Šantrůčková et al.,
2000; Kristiansen et al., 2004).
Carbon in exudation and root respiration derives mainly from C assimilated a few hours to days
ago (Craine et al., 1999; Högberg et al., 2001; Kuzyakov and Cheng, 2001). Moreover, the belowground and rhizosphere processes respond very rapidly to changes in nutrient supply and environmental conditions (Ekblad and Högberg, 2001; Kuzyakov and Cheng, 2001). Therefore, knowledge
20
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
of 13C fractionation of this recently assimilated C is especially important in studies based on natural
13
C labelling, which focus on fast, short-term processes such as root respiration and microbial respiration of root exudates. The balance of recently assimilated C as well as its contribution to belowground C fluxes can only be determined by pulse labelling of plant shoots in a 14CO2 atmosphere or
in an atmosphere which is enriched or depleted with 13CO2 compared to ambient air. In order to
determine the partitioning of recently assimilated C and the 13C fractionation under natural conditions, i.e. without any influences from artificial 13C labelling, we combined 14C pulse labelling with
13
C natural abundance.
Isotopic discrimination of 13C between the pools in a system strongly depends on the rate of
processes and partitioning of C allocated in each pool. Thus, the different extents of 13C discrimination observed in previous studies (Fischer and Tieszen, 1995; Scartazza et al., 2004) can be partly
linked with the C partitioning between above- and below-ground pools as well as between different
below-ground pools. The root-to-shoot ratio is strongly affected by plants’ requirements for nutrient
and water acquisition (Farrar and Jones, 2000; Andrews et al., 2001). Therefore, various nutrients
and especially N are controlling C partitioning between above- and below-ground pools (Merckx et
al., 1987; Liljeroth et al., 1990). Nitrogen fertilization is known to decrease the amount of C translocated into roots (Brown et al., 1996). We therefore used different levels of nutrient supply to control the C partitioning between above- and below-ground pools and to affect below-ground
processes. This allowed us to relate the 13C fractionation to the changes in pools and to recent
fluxes.
The specific objectives of this study were (1) to determine the amounts of recently assimilated C
in maize (Zea mays L.) shoots, roots, exudates, and CO2 derived from root respiration – in dependence of nutrient supply, (2) to estimate 13C fractionation of recently assimilated C, especially by
root respiration and exudation, and (3) to estimate the effect of different partitioning of recently
assimilated C on the 13C fractionation.
For these aims, maize shoots were pulse labelled three times in a 14CO2 atmosphere using 14C as
a tracer for recently assimilated C in the four pools: shoots, roots, exudates and CO2 from root respiration. In all four pools δ13C values were measured to determine fractionations between these
pools. As we hypothesized that 13C fractionation depends on C partitioning, three different levels of
nutrient supply were introduced to change C allocation pattern.
Materials and methods
Experimental set-up
Twenty-three maize plants (cv. Tassilo) were grown in standard nutrient solution under controlled
laboratory conditions, one plant per container. The maize grains were germinated on wet filter paper
first to a maximum leaf length of 5 cm. After six days the seedlings were transferred to 250 ml
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
21
polycarbonate filtration devices (SM16510/11, Sartorius, Göttingen, Germany), used here as pots
for plant growth. The pots were filled with the standard nutrient solution with full supply (K2SO4
153.4 mg l–1, KCl 7.5 mg l–1, Ca(NO3)2 · 4 H2O 472.3 mg l–1, MgSO4 · 7 H2O 246.5 mg l–1, KH2PO4
34.0 mg l–1, KNO3 303.3 mg l–1, H3BO3 0.0618 mg l–1, MnCl2 · 4 H2O 0.099 mg l–1, CuSO4 · 5 H2O
0.0499 mg l–1, (NH4)6Mo7O24 · 4 H2O 0.0247 mg l–1, ZnSO4 · 7 H2O 0.2875 mg l–1,
C10H12FeN2NaO8 aq. 36.7 mg l–1). Air was pumped through these pots from bottom to top with one
membrane pump (Type 113, Rietschle Thomas, Memmingen, Germany) connected by a tube to
every single pot. Nutrient solution was exchanged for fresh solution on day 11 of maize growth. On
day 14 after germination, the pots with the plants were sealed with non-phytotoxic silicone rubber
(TACOSIL 145, Thauer & Co., Dresden, Germany) between shoots and roots and the seal was
tested for air leaks. Another tube was connected from the top outlet of the filter devices to a CO2trapping tube filled with 20 ml 1 M NaOH solution. The output of the trapping tube was connected
to the input of the membrane pump. Thus, air containing CO2 evolved from root respiration circulated in a closed system: air was pumped through the nutrient solution, CO2 from root respiration
was trapped in NaOH solution, and the resulting CO2-free air coming out of these trapping tubes
was again pumped through the nutrient solution. A similar system is described in detail by Kuzyakov and Siniakina (2001). This circulation prevented contamination of the air inside the system by
atmospheric CO2 having a different δ13C.
Two hours after light on event on day 15, three treatments were started: full nutrient solution
(NS), 10 times diluted NS (DNS), and deionised water (DW). So, the full NS provided for all plants
before labelling was exchanged for one of the three treatments. The high-nutrient treatment consisted of 15 replications (to yield more biomass for further experiments) and the two low-nutrient
treatments consisted of four replications.
Labelling and sampling
On day 15 after germination the maize was labelled for the first time. All sealed pots with plants
were placed into a Plexiglas chamber (0.5 x 0.5 x 0.6 m³) for the labelling procedure described in
detail by Cheng et al. (1993). Briefly, the chamber was connected by tubing with a flask containing
2 ml 1 mM Na214CO3 solution to which 5 ml 9 M H2SO4 was added to produce 14CO2. The δ13C
value of air in the chamber and in the laboratory was assumed to be –8.1‰ (calculated from a δ13C
value of –7.8‰ in 1991 (Ehleringer, 1991; Boutton, 1996) and a progressive 13C reduction of
0.02‰ y–1 (Fung et al., 1997)). The plants were labelled during 1.5 hours in the atmosphere containing 5 MBq 14C and a concentration of 345 ppm atmospheric CO2 plus the amount of labelled
CO2. Usually, about 30 min of labelling time are required for C4 plants to reach the CO2 compensation point (Kuzyakov and Cheng, 2004). A longer time period was used in our experiment to increase the 14C incorporation into plant biomass. Before opening the labelling chamber, the chamber
air was pumped through 1 M NaOH solution to remove unassimilated 14CO2. Activities of unassimilated 14CO2 and of the 14C residue in the Na214CO3 source were subtracted from the total 14C,
22
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
which was in the flask before the start of the labelling, in order to calculate the total 14C input activity. The latter was divided by the number of plants in the labelling chamber resulting in an input
activity of 214.5 kBq per plant. After the labelling, the chamber was opened and the trapping of
CO2 evolved by root respiration was started.
The experiment consisted of three cycles started on days 14, 19, and 24. Each cycle included: (1)
supply of the plants with full NS for recovery from DNS or DW for one day before labelling (CO2
and exudates were collected during this period before second and third labelling), (2) 14C labelling
for 1.5 hours, and (3) trapping of CO2 in NaOH and of exudates released into NS, DNS, or DW for
four days. Due to nutrient solution uptake, the plants were provided with 300 ml instead of 250 ml
solution before the second and the third labelling. Plants were harvested on day 29 (after three cycles), divided into shoots and roots, dried at 40°C, and ground in a ball-mill. Immediately after
sampling of root exudates accumulated in the nutrient solutions, Micropur (Katadyn, Wallisellen,
Switzerland) containing Ag+ ions was added to the flask to suppress microbial decomposition of
exudates (Gransee and Wittenmayer, 2000; Kuzyakov and Siniakina, 2001) and the samples were
stored at 4°C before analysis.
Sample analyses
Total dissolved C released as exudates into NS was measured by a Dimatoc-100 TOC/TIC analyser
(Dimatec, Essen, Germany). C in shoots and roots was measured by a Euro EA C/N analyser
(EuroVector, Milan, Italy). CO2 trapped in NaOH solution during the sampling was precipitated
with 0.5 M BaCl2 solution and then the NaOH was titrated with 0.2 M HCl against phenolphthalein
indicator (Zibilske, 1994).
The 14C activity of 14CO2 trapped in NaOH solution or of exudates in nutrient solution was
measured in 2 ml aliquots added to 4 ml scintillation cocktail Rotiszint Eco Plus (Carl Roth,
Karlsruhe, Germany) after decay of chemiluminescence (for NaOH). 14C activity was measured
using a Wallac 1411 Liquid Scintillation Counter (Wallac Oy, Turku, Finland). The 14C counting
efficiency was about 87% and the 14C activity measurement error did not exceed 2%. The absolute
14
C activity was standardized by addition of NaOH solution as quencher to the scintillation cocktail
and using the spectrum of an external standard (SQP(E) method). 14C in solid samples (dried shoots
and roots) was measured after combustion of 200 mg of sample within an oxidizer unit (Model 307,
Canberra Packard Ltd., Meriden, USA), absorption of the 14C in Carbo-Sorb E (Perkin Elmer, Inc.,
Boston, USA), and addition of the scintillation cocktail Permafluor E+ (Perkin Elmer, Inc.).
For δ13C, 1 mg of ground maize shoots or roots was weighed out into tin capsules and analysed
on a Thermo Finnigan MAT DELTAplus Advantage isotope ratio mass spectrometer (IRMS from
Thermo Electron Corporation, Waltham, USA) coupled to the Euro EA C/N analyser. For δ13C
analysis of CO2 trapped in NaOH, an excess of 0.5 M BaCl2 solution was added to the NaOH trapping solution to form a precipitate of BaCO3. The BaCO3 precipitate was carefully washed ten
times with deionised water until pH of 7 was achieved. Washed BaCO3 was dried at 60°C and about
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
23
0.8 mg of dried BaCO3 were weighed out into tin capsules for δ13C analysis on IRMS. To prepare
exudates for IRMS analyses, 60 ml of nutrient solution containing exudates was dried at 60°C in a
Petri dish for each sample. The dry residue was scratched out of the dish with a steel spatula and
weighed out into tin capsules resulting in 25 µg exudates C per sample for IRMS analyses.
Statistical analyses
The experiment was conducted with four replicates for the low nutrient treatments and 15 replicates
for the high nutrient treatment. All replicates were analysed on 14C activity, C- and N-contents in
shoots, roots, exudates, and CO2. Only a choice of samples from each pool was analysed on δ13C
values (six, three, and four samples for NS, DNS, and DW treatments, respectively). 14C data are
presented as percentage of 14C recovered in shoots, roots, exudates, and CO2 after 29 days at the end
of the experiment. So, the 14C percentages are related to the 14C activity recovered after three labelling pulses. Standard deviation (SD) was calculated as a variability parameter.
Significance of differences between treatments was analysed for each sampling by one-way
ANOVA. We have calculated the least significant difference (LSD 0.05) in a post hoc NewmanKeuls test to identify differing treatments. Linear regressions were calculated between plant parts’
δ13C values and biomass and between the δ13C values of various plant parts. To relate δ13C values to
recently assimilated C, linear regressions between 14C (representing recently assimilated C) and
δ13C values were calculated. In the first step, the dependence between δ13C and 14C within each
treatment for each parameter was calculated. Since no significant regressions were found within the
treatments, we accepted that the variations of δ13C and 14C within each treatment were randomly
distributed and the means and standard deviations were calculated for each treatment. In the second
step, we calculated regressions between means of δ13C values (dependent variable) and means of
14
C (independent variable).
Results
Total carbon in different plant pools
The shoot dry matter of DNS and DW treatments was the same (Table 2-1). The value in the full
supply NS treatment was up to 1.8 g significantly higher than in the other two treatments (P <
0.001). There were also significant differences in the root mass between NS and DNS or DW treatments (P < 0.01), whereas root mass in the NS treatment was up to 0.3 g higher. The root dry matter
of DNS and DW treatments was the same. No significant differences between the three treatments
were found for the C content in shoots (391 mg g–1 on average). The C content in roots of DNS was
405 mg g–1 and was, at about 46 mg g–1, significantly higher (P < 0.05) than that of full NS (Table
2-1). No significant difference was found between shoots and roots of any treatment, this means
that the shoots and roots had the same C contents within each treatment.
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
24
Table 2-1: Dry matter, total C content, total N content, and C/N-ratio of 29-day-old maize (means ± SD).
Nutrient
solution
dry matter [g]
Shoots
Roots
total C [mg g-1]
Shoots
Roots
3.3 ±
0.9 ±
0.4
0.2
364.6 ± 28.9
358.5 ± 24.0
0.1 x nutrient
solution
1.8 ±
0.7 ±
0.3
0.1
406.1 ± 17.3
405.3 ± 11.4
Deionised
H2O
1.5 ±
0.6 ±
LSD
P (α = 0.05)
0.3
0.1
***
**
0.4
0.2
402.7 ± 26.5
390.1 ± 23.0
ns
*
41.3
34.2
total N [mg g-1]
Shoots
Roots
29.4 ±
20.3 ±
4.7
2.2
21.3 ±
16.1 ±
1.9
0.5
17.9 ±
16.9 ±
4.9
1.9
**
*
6.8
3.0
C/N
Shoots
Roots
12.7 ±
17.8 ±
2.4
1.8
19.1 ±
25.1 ±
0.9
1.4
23.6 ±
23.2 ±
5.8
2.4
**
***
5.7
3.0
*: P < 0.05, **: P < 0.01, ***: P < 0.001, ns: not significant
Contrasting to the total C content in shoots, there was a significantly lower total nitrogen (N)
content of 21 mg g–1 in DNS compared to 29 mg g–1 in full NS treatment (P < 0.01). Consequently,
the C/N ratio increased with decreasing nutrient supply from 13 for NS to 19 for DNS (P < 0.01).
The same effect of decreasing N content and increasing C/N with decreasing nutrient supply was
recorded for the roots (P < 0.05 for N; P < 0.001 for C/N). The N and C/N of shoots and of roots
also differed significantly between NS and DW treatments, but not between DNS and DW treatments. In all three treatments, the N content in the shoots was up to 9 mg g–1 higher than that of the
roots. Considering the equal C contents in shoots and roots, the C/N ratios of shoots and roots were
larger in DW than in NS. Thus, changing the nutrient supply affected the allocation of total N, but
not of total C, to shoots and roots.
The cumulative C exudation and the cumulative CO2 efflux were calculated from the first labelling to the end of the experiment. The C exudation in full NS was significantly higher (P < 0.001)
for all sampling dates compared to the other two treatments (Fig. 2-1a). The cumulative amount of
exuded C in full NS reached 111 ± 15 mg C per plant in 14 days, which was about four times higher
than in the nutrient-deficient treatments. No significant differences were found in the amount of
exuded C between DNS and DW treatments (Fig. 2-1a).
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
25
Fig. 2-1: Cumulative exudation (a) and cumulative root respiration (b) from maize between days 15 and 29
depending on growth media: ‘ full nutrient supply (NS), … 10 x diluted nutrient supply (DNS), and U deionised water (DW). Standard deviation shown to one side of the symbol only, least significant difference
(LSD, α = 0.05) shown for each sampling date in bars on top. Note different range of y-scale.
The cumulative CO2 efflux from NS treatment was 205 ± 45 mg C after 14 days and was always
higher compared to the other two treatments, which was significant on sampling days 24, 25, and 29
(P < 0.05). No significant differences were found in the cumulative CO2 efflux between treatments
on sampling days 19 and 20 (Fig. 2-1b).
Recently assimilated carbon (14C)
We labelled plants in a 14CO2 atmosphere to estimate the contribution of recently assimilated C to
root respiration and exudation, and we changed the nutrient supply to vary the partitioning of recently assimilated C. The total 14C recovery from shoots, roots, exudates, and CO2 from root respiration in relation to an input of 214.5 kBq per plant was not significantly different between the three
treatments. We found 44 to 53% 14C of the input per plant in all pools, i.e. the maize plants respired
47 to 56% of the 14C input via the shoots during 14 days after the first labelling. The distribution of
recently assimilated C of three-times labelled 29-day-old maize was: 76%, 73%, and 69% of recov-
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
26
ered 14C in the shoots and 8%, 10%, and 13% of recovered 14C in the roots for NS, DNS, and DW
treatments, respectively (Table 2-2). The amount of recovered 14C allocated in maize shoots was not
significantly different in the three treatments. The allocation of 14C in the roots was similar for DNS
and DW and for DNS and NS treatments, whereas the incorporation of 14C into roots of the NS
treatment was significantly lower than in the DW treatment (P < 0.05). Only 0.4%, 0.5%, and 0.8%
of recovered 14C were allocated to exudates, whereas 15%, 16%, and 17% of recovered 14C were
used for root respiration and were trapped as CO2 for NS, DNS, and DW treatments, respectively.
Recovery of 14C in exudates was the same for NS and DNS treatments. There was no significant
difference of 14C in CO2 of the three treatments.
Table 2-2: Distribution of 14C recovered in maize shoots, roots, exudates, and CO2 from root respiration after
three pulses of 14C labelling in relation to total 14C recovery per pot and sum of 14C activity in all four pools
in relation to 14C input per pot (means ± SD).
Nutrient
solution
C [% of 14C recovery]
Shoots
Roots
Exudates
root-derived CO2
0.1 x nutrient
solution
Deionised
H2O
LSD
P (α = 0.05)
14
C [% of 14C input]
sum of pools
±
±
±
±
5.8
2.5
0.1
4.6
52.8 ±
12.2
76.1
8.3
0.4
15.1
±
±
±
±
6.0
1.3
0.1
4.6
69.4
12.5
0.8
16.9
±
±
±
±
3.9
1.9
0.2
3.1
ns
*
**
ns
9.0
3.8
0.2
7.2
43.7 ±
0.2
45.8 ±
6.7
ns
17.4
73.0
10.1
0.5
16.4
14
*: P < 0.05, **: P < 0.01, ns: not significant
13
C discrimination depending on dry mass in shoots and roots and on δ13C in the
source compartment
Analysis of variance showed that the δ13C values of shoots were the same for DNS and DW treatments (–15.2‰; Table 2-3). These 13C-enriched values (by 0.2‰ compared to NS) were also observed for the whole plant. Shoots and roots δ13C values of the full NS treatment were significantly
lower (P < 0.05; P < 0.01; respectively) than those of the other two treatments. In all three treatments the δ13C value of the roots was higher than that of the shoots (P < 0.01 in the NS treatment).
The difference of δ13C values between shoots and roots increased from 0.3 to 0.6‰ with decreasing
nutrient concentration. Nutrient limitation led to smaller shoot and root dry mass, yielding higher
δ13C values in shoots (P < 0.05) and roots (P < 0.05), i.e. less 13C discrimination (Fig. 2-2, Table
2-4). These increasing δ13C values in the roots were significantly related to increasing δ13C values
in the shoots (Fig. 2-3a, Table 2-4).
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
27
Table 2-3: δ13C values of shoots, roots, exudates, and CO2 from root respiration from maize grown for 14
days in three different types of nutrient solution (means ± SD). Values for exudates and CO2 are shown for
three of five sampling times; values for shoots and roots are shown for 29-day-old maize.
13
δ C [‰]
0.1 x nutrient solution
Nutrient solution
Maize (29 days old)
Plants
Shoots
Roots
Exudates on days:
19
24
29
mean
CO2 on days:
19
24
29
mean
Differences for:
Roots - Shoots
Roots - Exudates
19
24
29
mean
Roots - CO2
19
24
29
mean
P
LSD
(α = 0.05)
Deionised H2O
-15.3 ±
-15.4 ±
-15.1 ±
0.1
0.1
0.1
-15.1 ±
-15.2 ±
-14.8 ±
0.0
0.1
0.1
-15.0 ±
-15.2 ±
-14.5 ±
0.1
0.0
0.3
***
*
**
0.1
0.2
0.3
-7.9
-9.9
-11.6
-9.8
±
±
±
±
1.1
2.6
1.4
1.9
-17.0
-15.8
-17.3
-16.7
±
±
±
±
0.4
0.3
0.8
0.8
-16.3
-16.2
-17.5
-16.7
±
±
±
±
0.7
0.2
1.0
0.8
***
*
**
***
1.6
3.7
2.8
2.5
-15.9
-16.1
-15.4
-15.8
±
±
±
±
0.6
0.5
0.2
0.3
-14.8
-14.4
-14.6
-14.6
±
±
±
±
0.9
0.1
0.6
0.2
-13.8
-14.2
-14.7
-14.2
±
±
±
±
0.3
0.3
0.5
0.5
***
***
*
**
1.0
0.6
0.8
0.7
0.1 **
0.3 ±
0.1
ns
0.6 ±
0.2
ns
1.7
1.7
3.0
2.1
±
±
±
±
0.6
0.3
0.9
0.7
ns
-0.7
-0.3
0.2
-0.3
±
±
±
±
0.3
0.3
0.5
0.5
ns
0.3 ±
-7.2
-5.2
-3.5
-5.3
±
±
±
±
0.9
2.1
1.1
1.6
***
**
***
***
2.2
1.0
2.4
1.9
±
±
±
±
0.3
0.3
0.7
0.7
ns
0.8
1.0
0.3
0.7
±
±
±
±
0.5
0.4
0.2
0.3
**
***
**
**
0.0
-0.5
-0.2
-0.2
±
±
±
±
0.7
0.1
0.5
0.2
ns
ns
*: P < 0.05, **: P < 0.01, ***: P < 0.001, : not significant
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
28
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
Fig. 2-2: δ13C values of (a) shoots, and (b) roots vs. dry mass of total maize plant parts (n = 13). Nutrient
concentrations are: ‘ full nutrient supply (NS), … 10 x diluted nutrient supply (DNS), and U deionised
water (DW). Note different range of x- and y-scales.
The δ13C value of exudates in the NS treatment collected from days 15 to 19 was at –7.9‰, significantly higher than the values of DNS and DW treatments (P < 0.001). Higher δ13C values of the
NS versus the other two treatments were also found for sampling days 24 (P < 0.05) and 29 (P <
0.01). Data for sampling days 20 and 25 are not shown, because all treatments provided full nutrient
supply one day before that sampling. Differences for these two sampling dates were not significant.
Comparing the mean δ13C values of exudates from sampling days 19, 24, and 29 with the δ13C
values of maize roots, significant differences were found only in the NS treatment (P < 0.001). The
δ13C value of exudates was 5.3‰ higher compared to maize roots (–15.1‰). For the DW and DNS
treatments, the δ13C values of exudates were 2.1 to 1.9‰ lower compared to –14.5 and –14.8‰ of
maize roots, but this was not significant. The relationship between δ13C in exudates and roots had a
significant R2 of 0.51 (Fig. 2-3b, Table 2-4).
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
29
Fig. 2-3: δ13C values of (a) roots vs. shoots (n = 13), (b) exudates vs. roots (n = 11), and (c) root-respired CO2
vs. roots of maize (n = 13). Nutrient concentrations are: ‘ full nutrient supply (NS), … 10 x diluted nutrient
supply (DNS), and U deionised water (DW). Note different range of x- and y-scales.
Significant differences were found between δ13C of CO2 evolved in NS and in DNS or DW
treatments (P < 0.05 to P < 0.001) on all sampling days. The δ13C value of CO2 respired by roots
grown in NS was 0.7‰ lower (P < 0.01) than the root value (–15.1‰). This significant difference
was recorded only for full nutrient supply. Increasing δ13C values in CO2 from root respiration were
very highly significantly related to increasing δ13C values in the roots (Fig. 2-3c, Table 2-4).
30
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
Table 2-4: Parameters of linear regressions of δ13C on plant dry mass (see Fig. 2-2); δ13C in roots on shoots,
exudates, and CO2 from root respiration (see Fig. 2-3); and δ13C on recently assimilated C (measured as 14C)
of maize shoots, roots, exudates and CO2 from root respiration for means of NS, DNS, and DW treatments
(see Tables 2-2 and 2-3).
Regression
variables
13
δ C-dry mass
shoots
roots
13
-15.01 ***
-14.20 ***
Slope
R²
-0.11 *
-0.95 *
0.43 *
0.35 *
1.51 *
-8.37 *
2.46 ***
0.44 *
0.51 *
0.81 ***
13
δ C-δ C
roots-shoots
exudates-roots
CO2-roots
13
Intercept
ns
8.14
-138.26 **
21.52 **
14
δ C- C
shoots
roots
exudates
CO2
-12.98 ns
-16.28 **
-7.55 ns
-29.21 *
-0.03 ns
0.14 *
-12.61 ns
0.89 *
0.82 ns
0.99 *
0.35 ns
0.99 *
*: P < 0.05, **: P < 0.01, ***: P < 0.001, ns: not significant
13
C discrimination of recently assimilated C
Despite a very high R2, the linear regression between recently assimilated C (14C) allocated to
shoots (Table 2-2) and their δ13C values (Table 2-3) was not significant (Table 2-4). For the roots a
significant relationship was comprised of decreasing nutrient supply, increasing amounts of recently
assimilated C, and increasing δ13C values (Tables 2-2, 2-3, and 2-4). Regression parameters of exudates δ13C on recently assimilated C were not significant (Table 2-4); thus, the former (Table 2-3)
was not linearly related with the latter (Table 2-2). Similarly to the roots, a significant relationship
was found for CO2 from root respiration (Table 2-4), i.e. with decreasing nutrient supply, the δ13C
value increased with increasing amounts of recently assimilated C (14C) (Tables 2-2 and 2-3).
Discussion
13
C fractionation of recently assimilated C in below-ground fluxes
Recently assimilated C by maize was traced by 14C labelling to quantify its balance in dependence
on differently concentrated nutrient solutions. With decreasing nutrient supply, 14C decreased in the
shoots and increased in the roots (Table 2-2). This confirms the frequently reported increase of C
allocation into the roots at decreasing nutrient availability (Merckx et al., 1987; Liljeroth et al.,
1990; Kuzyakov et al., 2002; Paponov and Engels, 2005) and supports the functional equilibrium
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
31
hypothesis of C allocation to the roots (Farrar and Jones, 2000). A lower translocation of recently
assimilated C into the roots at increased N fertilisation (Table 2-2) was overcompensated by the
higher total plant matter production of fertilised plants in our experiment (Table 2-1, compare
Kuzyakov et al. (2002)). Differences between the treatments in 14C of exudates and of CO2 were
significant only for the exudates, but 14C in root respiration tended to increase with decreasing nutrient supply (Table 2-2).
An allocation shift of recently assimilated C related to nutrient supply allowed us to investigate
the dependence of 13C isotopic fractionation on amounts of such C allocated to various plant parts
or C fluxes. The increase in shoot δ13C values from nutrient-rich to nutrient-poor conditions followed the slight decrease in recently assimilated C in the shoots (Tables 2-2 and 2-3). This increased translocation of photosynthate from shoots to roots under N stress may have altered δ13C in
shoots, independent of discrimination during carbon fixation (Brown et al., 1996). Although the
regression of δ13C vs. 14C in the shoots had a high R2 of 0.82, none of the regression parameters was
significant (Table 2-4). This absence of significant differences is connected with the fact that about
50% of the shoot mass was already present before the treatments with the three levels of nutrient
supply were started. Another explanation lies in the use of only three points for the regression calculation (see Tables 2-2 and 2-3), because the number of point pairs is crucial for the significance
level. Finally, a non-linearity of the relationship is possible.
We found a strong relationship between 14C and δ13C of maize roots: the higher the amount of
recently assimilated C translocated to the roots, the higher the root δ13C value (Tables 2-2 and 2-3).
Since δ13C values of roots and of CO2 from root respiration were significantly different only for the
NS treatment (Table 2-3), fractionation due to respiration cannot be considered to be the main process. The coefficient of determination allows us to conclude that the variance of root δ13C can be
entirely explained by the variance of recently assimilated 14C.
For the exudates, the regression between δ13C and 14C was not significant, i.e. we found no linear
relationship between the amount of recently assimilated C and 13C discrimination in exudates (Tables 2-2, 2-3, and 2-4). More data points are needed to identify a significant relationship, which
might be a non-linear one. The significant relationship for CO2 from root respiration can be described as follows: the more recently assimilated C was respired by the roots with decreasing nutrient supply, the more enriched in 13C was the CO2 (Tables 2-2, 2-3, and 2-4). This supports the
following hypothesis: the more C is translocated to a certain pool, the more enriched is that pool in
13
C.
Comparisons of δ13C values of plant pools and recently assimilated C (represented by 14C) were
only partly correct. The 14C labelled CO2 was assimilated by the shoots and was allocated to different plant pools. Thus, 14C in fact represented recently assimilated carbon. However, the δ13C values
of different plant pools (especially of the shoots and roots) were build not only by recently assimilated C, but also by carbon assimilated before the 14C labelling. Consequently, the δ13C value of
recently assimilated C contributes to the pools’ δ13C values, but does not represent them completely. Comparisons of allocation of recently assimilated C (measured as 14C) into the main photo-
32
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
synthetic products like sucrose or starch and their δ13C values could provide more specific and precise information to 13C fractionation. Such studies presume 13C compound specific analyses (Hobbie and Werner, 2004; Nogués et al., 2004), which were not available in our laboratory.
Dependence of 13C fractionation on biomass and C pools
A relationship between shoot δ13C values and shoot biomass has been frequently reported for wheat
(Condon et al., 1987; Araus et al., 1998; Fischer et al., 1998; Monneveux et al., 2005). In our experiment with maize, we have found the same significant relationship (Fig. 2-2a, Table 2-4). The
maize of our DNS and DW treatments produced less phytomass and showed higher shoot δ13C values compared to the full-supply NS treatment. Low nutrient concentrations, especially N, generally
result in a reduced water use efficiency (Stout and Schnabel, 1997; Caviglia and Sadras, 2001; Raven et al., 2004), which is linked to an increased stomatal conductance or a decreased net carbon
assimilation. However, nitrogen deficiency reduces stomatal conductance to some extent (Jacob et
al., 1990). Thus, a decrease of net carbon assimilation must be the reason for a low water use efficiency. The carbon assimilation between the three treatments was not significantly different, but a
trend of decreased assimilation by low nutrient supply could be seen. This decrease of assimilation
in the nutrient-poor treatments would have yielded a lower water use efficiency, a higher ratio of
intercellular space to atmospheric partial pressure of CO2 (pi/pa-ratio) and a higher 13C isotope discrimination by photosynthesis compared to nutrient-rich treatments. This increased discrimination,
i.e. more negative δ13C values, could not be found in our nutrient-poor treatments (Table 2-3). The
pi/pa-ratio was not measured in our experiment to check the internal CO2 concentration. Nor could
we calculate the pi/pa-ratio from shoot δ13C values since there is another unknown in C4 plant 13C
discrimination: φ , i.e. the proportion of carbon fixed by PEP carboxylase that subsequently leaks
out of the bundle sheath (Farquhar, 1983; Farquhar et al., 1989). Farquhar (1983) presented increased δ13C values in C4 plant shoots for a φ < 0.4. Thus, a small value φ representing the leakage
could compensate an increased pi/pa-ratio caused by nutrient deficiency and consequently lead to
more positive δ13C values in the shoots of nutrient-poor treatments. In order to determine the effect
of pi/pa or φ on
13
C discrimination in dependence of nutrient supply, pi/pa should be measured in
future observations. Furthermore, the photosynthetic carbon isotope discrimination in leaves towards air is needed to use the equation to calculate pi/pa or φ . Our results represent only the δ13C
values of shoots referred to the PDB standard. Since we did not measure δ13C of source air, we cannot calculate the discrimination in shoots referred to air CO2. The δ13C of source air should be determined in future observations to determine the photosynthetic carbon isotope discrimination.
Besides water use efficiency and leakage, Rubisco activity can also be influenced by altering nutrient supply. Ehleringer and Osmond (1989) reported a decrease in Rubisco activity at low leaf N
concentrations. This decrease in Rubisco activity should lead to an increased CO2 concentration in
the bundle sheath cells that might induce increased CO2 losses from this compartment by leaking.
According to the equation for C4 plants by Farquhar (1983), increased leakage in nutrient-poor
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
33
treatments of our experiment causes amplified 13C discrimination in the shoots. However, we observed a minor 13C discrimination in the shoots of nutrient-poor treatments than in the full NS
treatment (Table 2-3). Hence, low Rubisco activity and increased leakage cannot be the cause of the
13
C-enriched shoots of nutrient-poor treatments compared to the full NS treatment. Furthermore, a
reduced Rubisco activity in the nutrient-poor treatments is very unlikely since our plants were
grown in the same nutrient solution up to day 15 of the experiment and received full nutrient supply
one day before 14C labelling. Thus, a reduced Rubisco activity and an increased CO2 concentration
in the bundle sheath cells cannot be assumed for our nutrient-poor treatments compared to the NS
treatment.
Dependent on the C allocation pattern, isotopic fractionation of 13C occurred between shoots and
roots of maize; this was more pronounced for higher C allocation to roots versus shoots (Table 2-3).
Roots were significantly enriched in 13C compared to shoots, as reported in several other studies
(Tieszen and Boutton, 1989; Andreux et al., 1990; Fischer and Tieszen, 1995; Boutton, 1996; Hobbie et al., 2002; Badeck et al., 2005). Such 13C enrichments in the roots could arise by two plausible
post-photosynthetic fractionation mechanisms (Hobbie et al., 2004; Badeck et al., 2005): (1) the
formation of 13C-enriched carbohydrates, the primary transfer compounds in plants, or (2) discrimination against 13C during root respiration. Hobbie and Werner (2004) explained the former mechanism by the synthesis of 13C-depleted compounds in the leaves like lipids and lignin leading to 13Cenriched unreacted mobile sugars. These 13C-enriched sugars are then transported to the roots,
where again lipids and lignin are synthesized from these sugars. In this second fractionation lipids
and lignin are now more enriched then in the leaves due to the transport of the 13C-enriched substrate to the roots in contrast to the original substrate in the leaves.
In the NS treatment, exudates were between 4 and 7‰ enriched in 13C compared to roots (Table
2-3). This enrichment could be due to a fractionation of recently assimilated C between roots and
exudates. Hobbie et al. (2004) found a 2‰ isotopic enrichment of 13C in newly incorporated soil
carbon relative to Douglas-fir (Pseudotsuga mensiezii (Mirbel) Franco) root carbon. Cernusak et al.
(2005) reported an enrichment range of –0.3 to +4.6‰ in phloem sap sugars relative to Tasmanian
bluegum (Eucalyptus globulus Labill.) leaf dry matter. In a similar study on Triticum aestivum L.
Yoneyama et al. (1997) found an enrichment range of 1.4 to 1.6‰ in phloem relative to leaf blades.
Since exudates and phloem sap sugars both consist of recently assimilated C, the latter two studies
can also be compared to our one. Those three studies and our study show that 13C fractionation between roots and exudates occurs in a wide range for C3 and C4 plants and has to be further investigated for different species and nutrient supply. The extend of 13C fractionation between roots and
exudates could depend on the type of source compound used to produce exudates. Fractionation
could depend for example on the type of sucrose as one of the main exudation components. Tcherkez et al. (2004) gave some evidence that sucrose produced in the night from starch breakdown is
isotopically heavier than sucrose produced in the light from triose-phosphate.
Considering the δ13C of CO2 in the NS treatment, we found a significant depletion of about 0.7‰
compared to the roots (Table 2-3). Badeck et al. (2005) found a preferred release of the lighter iso-
34
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
tope relative to roots of kidney bean (Phaseolus vulgaris L.), leading to a 1.3‰ depletion of CO2 in
root respiration. Similar results were found in a study on common sunflower (Helianthus annuus
L.), perennial ryegrass (Lolium perenne L.), and alfalfa (Medicago sativa L.), where the respiratory
CO2 of whole roots was depleted by 0.5 to 5.4‰ relative to root tissues (Klumpp et al., 2005).
While the NS treatment showed a significant fractionation, the two nutrient-poor treatments showed
no significant fractionation between roots and CO2 from root respiration (Table 2-3). This means
that earlier results pointing to the absence of 13C fractionation by root respiration (Cerling et al.,
1991; Cheng, 1996; Amundson et al., 1998; Fu and Cheng, 2002) must be interpreted with caution
and related to the C allocation pattern. 13C fractionation during root respiration could also be a
cause of roots’ 13C enrichment in contrast to the shoots as suggested by Badeck et al. (2005).
Conclusions
A separate evaluation of the effects of C pools and of recent C fluxes on the fractionation of 13C in
shoots, roots, exudates and root respiration was possible by a unique coupling of 14C tracing after
pulse labelling with measurements of δ13C values. Roots and root-respired CO2 were less depleted
in 13C when increasing amounts of recently assimilated C were translocated to these pools. This
relationship shows that the allocation of recently assimilated C, which depends on nutrient supply
and changing environmental conditions, may be an important factor controlling δ13C fractionation
within the plant–soil system. Therefore, the allocation pattern of assimilated C within the plant
should be considered for corrections of 13C fractionation in studies based on 13C natural abundance.
Acknowledgements
The German Research Foundation (DFG) supported this study. The authors would also like to thank
Dr. V. Cercasov for usage of the scintillation counter and Dr. W. Armbruster for the IRMS analyses.
References
Amundson, R., Stern, L., Baisden, T., Wang, Y., 1998. The isotopic composition of soil and soilrespired CO2. Geoderma 82, 83-114.
Andreux, F., Cerri, C.C., Vose, P.B., Vitorello, V.A., 1990. Potential of stable isotope, 15N and
13
C, methods for determining input and turnover in soils. In: Harrison, A.F., Ineson, P., Heal,
O.W. (Eds.), Nutrient cycling in terrestrial ecosystems, Elsevier Applied Sciences, New
York, pp. 259-275.
Andrews, M., Raven, J.A., Sprent, J.I., 2001. Environmental effects on dry matter partitioning
between shoot and root of crop plants: relations with growth and shoot protein concentration.
Annals of Applied Biology 138, 57-68.
Araus, J.L., Amaro, T., Casadesus, J., Asbati, A., Nachit, M.M., 1998. Relationship between ash
content, carbon isotope discrimination and yield in durum wheat. Australian Journal of
Plant Physiology 25, 835-842.
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
35
Badeck, F.-W., Tcherkez, G., Nogués, S., Piel, C., Ghashghaie, J., 2005. Post-photosynthetic
fractionation of stable carbon isotopes between plant organs – a widespread phenomenon.
Rapid Communications in Mass Spectrometry 19, 1381-1391.
Balesdent, J., Mariotti, A., Guillet, B., 1987. Natural 13C abundance as a tracer for studies of soil
organic matter dynamics. Soil Biology & Biochemistry 19, 25-30.
Barber, D.A., Martin, J.K., 1976. The release of organic substances by cereal roots in soil. New
Phytologist 76, 69-80.
Boutton, T.W., 1996. Stable carbon isotope ratios of soil organic matter and their use as indicators
of vegetation and climate change. In: Boutton, T.W., Yamasaki, S.I. (Eds.), Mass
Spectrometry of Soils, Marcel Dekker, New York, pp. 47-82.
Brown, K.R., Thompson, W.A., Camm, E.L., Hawkins, B.J., Guy, R.D., 1996. Effects of N addition rates on the productivity of Picea sitchensis, Thuja plicata, and Tsuga heterophylla
seedlings. II. Photosynthesis, 13C discrimination and N partitioning in foliage. Trees 10, 198205.
Caviglia, O.P., Sadras, V.O., 2001. Effect of nitrogen supply on crop conductance, water- and
radiation-use efficiency of wheat. Field Crops Research 69, 259-266.
Cerling, T.E., Solomon, D.K., Quade, J., Bowman, J.R., 1991. On the isotopic composition of
carbon in soil carbon dioxide. Geochimica et Cosmochimica Acta 55, 3404-3405.
Cernusak, L.A., Farquhar, G.D., Pate, J.S., 2005. Environmental and physiological controls over
oxygen and carbon isotope composition of Tasmanian blue gum, Eucalyptus globulus. Tree
Physiology 25, 129-146.
Cheng, W., 1996. Measurement of rhizosphere respiration and organic matter decomposition using
natural 13C. Plant and Soil 183, 263-268.
Cheng, W., Johnson, D.W., Fu, S., 2003. Rhizosphere effects on decomposition: controls of plant
species, phenology, and fertilization. Soil Science Society of America Journal 67, 14181427.
Cheng, W., Coleman, D.C., Carroll, C.R., Hoffman, C.A., 1993. In situ measurement of root
respiration and soluble C concentrations in the rhizosphere. Soil Biology & Biochemistry 25,
1189-1196.
Condon, A.G., Richards, R.A., Farquhar, G.D., 1987. Carbon isotope discrimination is positively
correlated with grain yield and dry matter production in field-grown wheat. Crop Science 27,
996-1001.
Craine, J.M., Wedin, D.A., Chapin, F.S.I., 1999. Predominance of ecophysiological controls on
soil CO2 flux in a Minnesota grassland. Plant and Soil 207, 77-86.
Duranceau, M., Ghashghaie, J., Badeck, F., Deléens, E., Cornic, G., 1999. δ13C of CO2 respired
in the dark in relation to δ13C of leaf carbohydrates in Phaseolus vulgaris L. under progressive drought. Plant, Cell and Environment 22, 515-523.
Ehleringer, J.R., 1991. 13C/12C fractionation and its utility in terrestrial plant studies. In: Coleman,
D.C., Fry, B. (Eds.), Carbon Isotope Techniques. Isotopic techniques in plant, soil, and
aquatic biology, Academic Press, Inc., San Diego, pp. 187-200.
Ehleringer, J.R., Osmond, C.B., 1989. Stable isotopes. In: Pearcy, R.W., Ehleringer, J.R.,
Mooney, H.A., Rundel, P.W. (Eds.), Plant Physiological Ecology, Chapman and Hall, London, pp. 281-300.
Ekblad, A., Högberg, P., 2001. Natural abundance of 13C in CO2 respired from forest soils reveals
speed of link between tree photosynthesis and root respiration. Oecologia 127, 305-308.
Farquhar, G.D., 1983. On the nature of carbon isotope discrimination in C4 species. Australian
Journal of Plant Physiology 10, 205-226.
36
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
Farquhar, G.D., Ehleringer, J.R., Hubick, K.T., 1989. Carbon isotope discrimination and photosynthesis. Annual Review of Plant Physiology and Plant Molecular Biology 40, 503-537.
Farrar, J.F., Jones, D.L., 2000. The control of carbon acquisition by roots. New Phytologist 147,
43-53.
Fischer, J.C.v., Tieszen, L.L., 1995. Carbon isotope characterization of vegetation and soil organic
matter in subtropical forests in Luquillo, Puerto Rico. Biotropica 27, 138-148.
Fischer, R.A., Rees, D., Sayre, K.D., Lu, Z.M., Condon, A.G., Larqué-Saavedra, A., 1998.
Wheat yield progress associated with stomatal conductance and photosynthetic rates, and
cooler canopies. Crop Science 38, 1467-1475.
Fu, S., Cheng, W., 2002. Rhizosphere priming effects on the decomposition of soil organic matter
in C4 and C3 grassland soils. Plant and Soil 238, 289-294.
Fung, I., Field, C.B., Berry, J.A., Thompson, M.V., Randerson, J.T., Malmström, C.M., Vitousek, P.M., Collatz, G.J., Sellers, P.J., Randall, D.A., Denning, A.S., Badeck, F., John,
J., 1997. Carbon 13 exchanges between the atmosphere and biosphere. Global Biogeochemical Cycles 11, 507-533.
Gerzabek, M.H., Haberhauer, G., Kirchmann, H., 2001. Soil organic matter pools and carbon13 natural abundances in particle-size fractions of a long-term agricultural field experiment
receiving organic amendments. Soil Science Society of America Journal 65, 352-358.
Ghashghaie, J., Duranceau, M., Badeck, F.-W., Cornic, G., Adeline, M.-T., Deléens, E., 2001.
δ13C of CO2 respired in the dark in relation to δ13C of leaf metabolites: comparison between
Nicotiana sylvestris and Helianthus annuus under drought. Plant, Cell and Environment 24,
505-515.
Ghashghaie, J., Badeck, F.-W., Lanigan, G., Nogués, S., Tcherkez, G., Deléens, E., Cornic, G.,
Griffiths, H., 2003. Carbon isotope fractionation during dark respiration and photorespiration in C3 plants. Phytochemistry Reviews 2, 145-161.
Gransee, A., Wittenmayer, L., 2000. Qualitative and quantitative analysis of water-soluble root
exudates in relation to plant species and development. Journal of Plant Nutrition and Soil
Science 163, 381-385.
Hobbie, E.A., Werner, R.A., 2004. Intramolecular, compound-specific, and bulk carbon isotope
patterns in C3 and C4 plants: a review and synthesis. New Phytologist 161, 371-385.
Hobbie, E.A., Tingey, D.T., Rygiewicz, P.T., Johnson, M.G., Olszyk, D.M., 2002. Contributions
of current year photosynthate to fine roots estimated using a 13C-depleted CO2 source. Plant
and Soil 247, 233-242.
Hobbie, E.A., Johnson, M.G., Rygiewicz, P.T., Tingey, D.T., Olszyk, D.M., 2004. Isotopic estimates of new carbon inputs into litter and soils in a four-year climate change experiment
with Douglas-fir. Plant and Soil 259, 331-343.
Högberg, P., Nordgren, A., Buchmann, N., Taylor, A.F.S., Ekblad, A., Högberg, M.N., Nyberg, G., Ottosson-Löfvenius, M., Read, D.J., 2001. Large-scale forest girdling shows that
current photosynthesis drives soil respiration. Nature 411, 789-792.
Jacob, J., Udayakumar, M., Prasad, T.G., 1990. Mesophyll conductance was inhibited more than
stomatal conductance in nitrogen deficient plants. Plant Physiology and Biochemistry-New
Delhi 17, 55-61.
John, B., Ludwig, B., Flessa, H., 2003. Carbon dynamics determined by natural 13C abundance in
microcosm experiments with soils from long-term maize and rye monocultures. Soil Biology
& Biochemistry 35, 1193-1202.
Klumpp, K., Schäufele, R., Lötscher, M., A., L.F., Feneis, W., Schnyder, H., 2005. C-isotope
composition of CO2 respired by shoots and roots: fractionation during dark respiration?
Plant, Cell and Environment 28, 241-250.
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
37
Kristiansen, S.M., Brandt, M., Hansen, E.M., Magid, J., Christensen, B.T., 2004. 13C signature
of CO2 evolved from incubated maize residues and maize-derived sheep faeces. Soil Biology
& Biochemistry 36, 99-105.
Kuzyakov, Y., 2001. Tracer studies of carbon translocation by plants from the atmosphere into the
soil (A Review). Eurasian Soil Science 34, 28-42.
Kuzyakov, Y., 2002. Separating microbial respiration of exudates from root respiration in nonsterile soils: a comparison of four methods. Soil Biology & Biochemistry 34, 1621-1631.
Kuzyakov, Y., Domanski, G., 2000. Carbon input by plants into the soil. Review. Journal of Plant
Nutrition and Soil Science 163, 421-431.
Kuzyakov, Y., Siniakina, S.V., 2001. A novel method for separating root-derived organic compounds from root respiration in non-sterilized soils. Journal of Plant Nutrition and Soil Science 164, 511-517.
Kuzyakov, Y., Cheng, W., 2001. Photosynthesis controls of rhizosphere respiration and organic
matter decomposition. Soil Biology & Biochemistry 33, 1915-1925.
Kuzyakov, Y., Cheng, W., 2004. Photosynthesis controls of CO2 efflux from maize rhizosphere.
Plant and Soil 263, 85-99.
Kuzyakov, Y., Kretzschmar, A., Stahr, K., 1999. Contribution of Lolium perenne rhizodeposition
to carbon turnover of pasture soil. Plant and Soil 213, 127-136.
Kuzyakov, Y., Siniakina, S.V., Ruehlmann, J., Domanski, G., Stahr, K., 2002. Effect of nitrogen fertilisation on below-ground carbon allocation in lettuce. Journal of the Science of
Food and Agriculture 82, 1432-1441.
Liljeroth, E., van Veen, J.A., Miller, H.J., 1990. Assimilate translocation to the rhizosphere of
two wheat lines and subsequent utilization by rhizosphere microorganisms at two soil nitrogen concentrations. Soil Biology & Biochemistry 22, 1015-1021.
Liljeroth, E., Kuikman, P., van Veen, J.A., 1994. Carbon translocation to the rhizosphere of
maize and wheat and influence on the turnover of native soil organic matter at different soil
nitrogen levels. Plant and Soil 161, 233-240.
Mary, B., Mariotti, A., Morel, J.L., 1992. Use of 13C variations at natural abundance for studying
the biodegradation of root mucilage, roots and glucose in soil. Soil Biology & Biochemistry
24, 1065-1072.
Merckx, R., Dijkstra, A., den Hartog, A., van Veen, J.A., 1987. Production of root-derived material and associated microbial growth in soil at different nutrient levels. Biology and Fertility
of Soils 5, 126-132.
Monneveux, P., Reynolds, M.P., Trethowan, R., González-Santoyo, H., Peña, R.J., Zapata, F.,
2005. Relationship between grain yield and carbon isotope discrimination in bread wheat
under four water regimes. European Journal of Agronomy 22, 231-242.
Nogués, S., Tcherkez, G., Cornic, G., Ghashghaie, J., 2004. Respiratory carbon metabolism following illumination in intact French bean leaves using 13C/12C isotope labeling. Plant Physiology 136, 3245-3254.
Paponov, I.A., Engels, C., 2005. Effect of nitrogen supply on carbon and nitrogen partitioning after flowering in maize. Journal of Plant Nutrition and Soil Science 168, 447-453.
Raven, J.A., Handley, L.L., Wollenweber, B., 2004. Plant nutrition and water use efficiency. In:
Bacon, M.A. (Ed.) Water Use Efficiency in Plant Biology. Biological Sciences Series,
Blackwell Publishing Ltd., Oxford, pp. 171-197.
Rochette, P., Flanagan, L.B., 1997. Quantifying rhizosphere respiration in a corn crop under field
conditions. Soil Science Society of America Journal 61, 466-474.
38
2 Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize
Rochette, P., Angers, D.A., Flanagan, L.B., 1999. Maize residue decomposition measurement
using soil surface carbon dioxide fluxes and natural abundance of carbon-13. Soil Science
Society of America Journal 63, 1385-1396.
Šantrůčková, H., Bird, M.I., Frouz, J., Šustr, V., Tajovský, K., 2000. Natural abundance of 13C
in leaf litter as related to feeding activity of soil invertebrates and microbial mineralisation.
Soil Biology & Biochemistry 32, 1793-1797.
Scartazza, A., Mata, C., Matteucci, G., Yakir, D., Moscatello, S., Brugnoli, E., 2004. Comparisons of δ13C of photosynthetic products and ecosystem respiratory CO2 and their responses
to seasonal climate variability. Oecologia 140, 340-351.
Søe, A.R.B., Giesemann, A., Anderson, T.H., Weigel, H.-J., Buchmann, N., 2004. Soil respiration under elevated CO2 and its partitioning into recently assimilated and older carbon
sources. Plant and Soil 262, 85-94.
Stewart, D.P.C., Metherell, A.K., 1999. Carbon (13C) uptake and allocation in pasture plants following field pulse-labelling. Plant and Soil 210, 61-73.
Stout, W.L., Schnabel, R.R., 1997. Water use efficiency of perennial rye grass as affected by soil
drainage and nitrogen fertilization on two floodplain soils. Journal of Soil and Water Conservation 52, 207-211.
Swinnen, J., van Veen, J.A., Merckx, R., 1994a. 14C pulse-labelling of field-grown spring wheat:
an evaluation of its use in rhizosphere carbon budget estimations. Soil Biology & Biochemistry 26, 161-170.
Swinnen, J., van Veen, J.A., Merckx, R., 1994b. Rhizosphere carbon fluxes in field-grown spring
wheat: model calculations based on 14C partitioning after pulse-labeling. Soil Biology & Biochemistry 26, 171-182.
Tcherkez, G., Farquhar, G., Badeck, F., Ghashghaie, J., 2004. Theoretical considerations about
carbon isotope distribution in glucose of C3 plants. Functional Plant Biology 31, 857-877.
Tcherkez, G., Nogués, S., Bleton, J., Cornic, G., Badeck, F., Ghashghaie, J., 2003. Metabolic
origin of carbon isotope composition of leaf dark-respired CO2 in French bean. Plant Physiology 131, 237-244.
Tieszen, L.L., Boutton, T.W., 1989. Stable carbon isotopes in terrestrial ecosystem research. In:
Rundel, P.W., Ehleringer, J.R., Nagy, K.A. (Eds.), Stable isotopes in ecological research,
Springer, New York, pp. 167-195.
Warembourg, F.R., Estelrich, D.H., 2000. Towards a better understanding of carbon flow in the
rhizosphere: a time-dependent approach using carbon-14. Biology and Fertility of Soils 30,
528-534.
Whipps, J.M., Lynch, J.M., 1983. Substrate flow and utilization in the rhizosphere of cereals. New
Phytologist 95, 605-623.
Yoneyama, T., Handley, L.L., Scrimgeour, C.M., Fisher, D.B., Raven, J.A., 1997. Variations of
the natural abundances of nitrogen and carbon isotopes in Triticum aestivum, with special
reference to phloem and xylem exudates. New Phytologist 137, 205-213.
Zibilske, L.M., 1994. Carbon Mineralization. In: Weaver, R.W., Angle, S., Bottomley, P., Bezdicek, D., Smith, S., Tabatabai, A., Wollum, A. (Eds.), Methods of Soil Analysis, Part 2,
Microbiological and Biochemical Properties. Soil Science Society of America Book Series,
No. 5, Soil Sci. Soc. Am., Inc., Madison, pp. 835-864.
3
Three-source partitioning of CO2 efflux from soil planted with
maize by 13C natural abundance fails due to inactive microbial
biomass
Martin Werth, Irina Subbottina and Yakov Kuzyakov
Soil Biology & Biochemistry (2006), 38: 2772-2781
includes one table and six figures
with kind permission of Elsevier Science Ltd
40
3 Three-source partitioning of CO2 efflux from soil planted with maize
Abstract
A theoretical approach to the partitioning of carbon dioxide (CO2) efflux from soil with a C3
vegetation history planted with maize (Zea mays), a C4 plant, into three sources, root respiration
(RR), rhizomicrobial respiration (RMR), and microbial soil organic matter (SOM) decomposition
(SOMD), was examined. The δ13C values of SOM, roots, microbial biomass, and total CO2 efflux
were measured during a 40-day growing period. A three-source isotopic mass balance based on the
measured δ13C values and on assumptions made in other studies showed that RR, RMR, and SOMD
amounted 91%, 4%, and 5%, respectively. Two assumptions were thoroughly examined in a sensitivity analysis: the absence of 13C fractionation, and the conformity of δ13C of microbial CO2 and
that of microbial biomass. This approach strongly overestimated RR and underestimated RMR and
microbial SOM decomposition. Carbon dioxide efflux from unplanted soil was enriched in 13C by
2.0‰ compared to microbial biomass. The consideration of this 13C fractionation in the mass balance equation changed the proportions of RR and RMR by only 4% and did not affect SOMD. A
calculated δ13C value of microbial CO2 by a mass balance equation including active and inactive
parts of microbial biomass was used to adjust a hypothetical below-ground CO2 partitioning to the
measured and literature data. The active microbial biomass in the rhizosphere amounted to 37% to
achieve an appropriate ratio between RR and RMR compared to measured data. Therefore, the
three-source partitioning approach failed due to a low active portion of microbial biomass, which is
the main microbial CO2 source controlling the δ13C value of total microbial biomass. Since fumigation-extraction reflects total microbial biomass, its δ13C value was unsuitable to predict δ13C of released microbial CO2 after a C3–C4 vegetation change. The second adjustment to the CO2
partitioning results in the literature showed that at least 71% of the active microbial biomass utilizing maize rhizodeposits would be necessary to achieve that proportion between RR and RMR observed by other approaches based on 14C labelling. The method for partitioning total below-ground
CO2 efflux into three sources using a natural 13C labelling technique failed due to the small proportion of active microbial biomass in the rhizosphere. This small active fraction led to a discrepancy
between δ13C values of microbial biomass and of microbially respired CO2.
Abbreviations:
SOM – soil organic matter, RR – root respiration, RMR – rhizomicrobial
respiration, SOMD – soil organic matter decomposition
3 Three-source partitioning of CO2 efflux from soil planted with maize
41
Introduction
Partitioning the total carbon dioxide (CO2) efflux from soil is very important to identify individual
sinks or sources of CO2. Root-derived CO2 and CO2 derived from soil organic matter (SOM) decomposition can be quantified by isotopic labelling of plants with 13C or 14C isotopes and tracing
the label in root-derived CO2 (Ekblad and Högberg, 2001; Kuzyakov and Cheng, 2001). The difference between this labelled fraction and total CO2 efflux represents CO2 from soil organic matter
decomposition (SOMD). The above studies on below-ground CO2 from a boreal forest dominated
by Pinus sylvestris and Vaccinium myrtillus (Ekblad and Högberg, 2001) and from soil planted with
wheat (Triticum aestivum) (Kuzyakov and Cheng, 2001) revealed that about 70% of the CO2 was
derived from rhizosphere respiration and 30% from SOMD. However, these values vary strongly
dependent on plants, soils, and environmental conditions. It is exceptionally difficult to further differentiate between CO2 which is directly derived from root respiration and that which is derived
from mineralisation of rhizodeposits (Killham and Yeomans, 2001). This separation of root respiration (RR) and rhizomicrobial respiration (RMR) is one of the greatest challenges to quantifying
rhizosphere carbon flows. Separation is important to quantify carbon sources for SOM and for
rhizosphere microorganisms, to identify respiration of autotrophic and heterotrophic organisms, and
to calculate carbon turnover by rhizosphere microorganisms (Kuzyakov, 2004).
To date, five adequate methods have been suggested to separate RR and RMR in non-sterile
soils:
1) the isotope dilution method (Cheng et al., 1993), i.e. isotopic dilution of rhizomicrobial
14
CO2 by addition of unlabelled glucose to the rhizosphere of 14C labelled plants, where
14
CO2 from RMR is inversely proportional to the glucose concentration in the rhizosphere,
whereas 14CO2 of RR is not affected by glucose addition,
2) the model rhizodeposition technique (Swinnen, 1994), where two variants are used: (1) 14C
pulse labelled plants without model rhizodeposits (RR and RMR), and (2) 14C labelled
model rhizodeposits (glucose or plant extracts) added to soil with unlabeled plants (RMR),
3) modelling of 14CO2 efflux dynamics (Kuzyakov et al., 1999; Kuzyakov et al., 2001; Kuzyakov and Domanski, 2002), where a mathematical model is used to split up the curve of
14
CO2 efflux from soil with 14C labelled plants into RR and RMR by temporal delay of rhizomicrobial 14CO2 compared to 14CO2 from root respiration,
4) the exudate elution procedure (Kuzyakov and Siniakina, 2001), based on the rapid elution of
14
C labelled exudates from soil before microorganisms utilize them,
5) the difference method between root-derived 14CO2 and rhizomicrobial 14CO2 (Johansson,
1992), where root-derived 14CO2 evolved from the rhizosphere of plants continuously labelled in a 14CO2 atmosphere (RR and RMR) is compared with 14CO2 evolved by decomposition of uniformly 14C labelled rhizodeposits (RMR) obtained from the same plants.
42
3 Three-source partitioning of CO2 efflux from soil planted with maize
These methods, their basic assumptions, as well as possible error sources have been described in
detail earlier (Kuzyakov, 2002; Kuzyakov and Larionova, 2005). The first four methods are based
on pulse labelling of shoots in a 14CO2 atmosphere and subsequent monitoring of 14CO2 efflux from
the soil. However, the basic assumptions and principles of these methods, as well as the results observed in the original papers, all differ from one another. The comparison of the first four methods
in a single experiment under equal conditions showed that 14CO2 efflux from ryegrass (Lolium perenne) rhizosphere grown on a loamy Haplic Luvisol consisted of 40 – 50% RR and of about 50 –
60% RMR (Kuzyakov, 2002). The comparison showed that the isotope dilution method (Cheng et
al., 1993) and the method based on modelling 14CO2 efflux dynamics (Kuzyakov et al., 1999;
Kuzyakov et al., 2001; Kuzyakov and Domanski, 2002) are the most reliable methods, because they
showed similar separation results despite mutually exclusive assumptions. In the former method the
ratio of 14C in CO2 from root respiration to that derived from microbial respiration of rhizodeposits
is assumed to be constant during the observation, whereas this ratio is variable in the latter method.
Component integration (Edwards and Harris, 1977) and tree girdling (Högberg et al., 2001) are
two further methods, which were tested to separate RR and RMR. Their shortcomings, including
non-comparable respiration rates of disturbed and undisturbed soil in component integration or
stopping of RR and RMR by tree girdling, are discussed in detail by Kuzyakov (2005). Due to
many difficulties and non-testable assumptions, none of the suggested methods is acceptable as a
standard procedure for separately estimating RR and RMR. Owing to these uncertainties, new and
more reliable approaches are required to separate root, rhizomicrobial, and SOM respiration types.
The objective of this study was to verify an approach to a quantitative estimation of (1) root respiration, (2) rhizomicrobial respiration, and (3) microbial respiration from SOM decomposition in
non-sterile soils. The theoretical approach was recently suggested by Kuzyakov (2004; 2005) and
was practically tested here. The method is based on the natural 13C labelling technique (Balesdent
and Mariotti, 1996), i.e. 13C natural abundance is used by growing C4 plants on a soil developed
under C3 vegetation (‘C3 soil’) or vice versa. Hence, the δ13C values of SOM, maize roots, microbial biomass, and total CO2 efflux from the soil are used to determine the three fractions of CO2.
These contributions of RR, RMR, and SOMD to total soil CO2 efflux can be calculated according to
the isotopic mass balance of microbial biomass and CO2. This method involves two assumptions
concerning 13C isotopic effects during root and microbial respiration:
1) The δ13C isotope signature of CO2 from rhizosphere respiration is the same as the δ13C value
of the roots.
2) The δ13C isotope signature of CO2 respired by microorganisms corresponds to the δ13C value
of microbial biomass.
A verification and discussion of these assumptions is provided here.
3 Three-source partitioning of CO2 efflux from soil planted with maize
43
Materials and methods
Experimental set-up
Twenty maize plants (Zea mays L.) were grown under controlled laboratory conditions on a loamy
Haplic Luvisol from loess with C3 vegetation history (Lolium perenne L.), collected from the University of Hohenheim’s research farm ‘Heidfeldhof’ in Stuttgart, Germany. The maize seeds (cv.
Tassilo) were germinated on wet filter paper. One day after germination the seedlings were transferred to 250 ml polycarbonate filtration devices (SM16510/11, Sartorius, Germany) filled with 400
g of the C3 soil, one plant per container (Fig. 3-1). A control treatment with one unplanted pot per
sampling date was established, which was treated exactly in the same way like the planted treatment. One day before the start of CO2 trapping, the holes in the pots around the plant shoots were
sealed with silicone rubber (TACOSIL 145, Thauer & Co., Germany) between roots and shoots,
and the seal was tested for air leaks. Trapping of CO2 from soil air started on day 9 after germination in a closed system for each plant (or control treatment). Air was pumped through every single
pot from bottom to top by a membrane pump (Type 113, Rietschle Thomas, Germany, pumping
rate 100 ml min–1), which was connected to the pot by a tube (Fig. 3-1). Another tube was connected to the top outlet of the filter device and to a CO2 trapping tube filled with 20 ml 1 M sodium
hydroxide (NaOH) solution. The output of the trapping tube was connected to the input of the
membrane pump. Therefore, the air containing CO2 evolved from soil respiration circulated in a
closed system. Firstly, the air was pumped through the pot, with any CO2 from total soil respiration
being trapped in NaOH solution. Secondly, the resulting CO2-free air coming from the NaOH trapping tube was pumped back through the pot. Thus, the air cycling was closed and was done continuously by the membrane pump.
The soil moisture was maintained gravimetrically at about 25% of the water-holding capacity
throughout the experiment by controlling the pots’ weights after the first water addition. On days 9,
15, 21, 27, and 33 after germination, a full fertilizer (5 kg nitrate-N ha–1, 0.4 kg monophosphate-P
ha–1, 10 kg K+ ha–1; see Werth and Kuzyakov (2005) for further detail) was added with the water to
the soil from one to five times depending on the date of sampling of the pots.
44
3 Three-source partitioning of CO2 efflux from soil planted with maize
Silicone
rubber
NaOH
Soil
Rubber
rings
Filter
support
Membrane
pump
Fig. 3-1: Experimental set-up for trapping of below-ground CO2 in NaOH solution. White arrows
show air flow.
Sampling and analyses
Soil and plants were destructively sampled in four replicates (i.e. one replicate for the control treatment) on days 16, 22, 28, 34, and 40 after germination. At harvest, each shoot was cut at the base,
the lid of the pot was opened and each root–soil column pulled out of the pot. The soil was divided
into bulk soil, rhizosphere and non-rhizosphere soil. Bulk soil was sampled by cutting a small
wedge into the soil column from the edge towards the centre. We then loosened the soil column
from the edge to gain the non-rhizosphere fraction. The soil adhering to the roots was collected as
the rhizosphere fraction. Only the results of the rhizosphere fraction are presented here. The moist
soil samples were immediately frozen until preparation for microbial biomass was started. The roots
were carefully washed with deionised water to remove soil particles. Shoots and roots were dried at
3 Three-source partitioning of CO2 efflux from soil planted with maize
45
40°C. Carbon dioxide trapped in NaOH was sampled on the harvest days and additionally once to
twice between two harvest days.
To estimate total CO2 efflux, the CO2 trapped in NaOH solution was precipitated with 0.5 M barium chloride (BaCl2) solution and then the NaOH was titrated with 0.2 M hydrochloric acid (HCl)
against phenolphthalein indicator (Zibilske, 1994). Soil microbial biomass was determined by the
chloroform fumigation extraction method (modified after Vance et al. (1987)). Roots were removed
from the unfrozen soil by handpicking and 10 g of soil were extracted with 40 ml of 0.05 M potassium sulphate (K2SO4) solution. Another 10 g of soil were firstly fumigated with chloroform for 24
h and then extracted in the same way. The K2SO4 and soil mixtures were shaken for 1 h at 200 rev
min–1, centrifuged at 3000 rev min–1 for 10 min, and then filtrated through a ceramic vacuum filter.
The extracts were frozen until analyses for total carbon (C) and nitrogen (N) concentrations were
done with a Dimatoc-100 TOC/TIC analyser (Dimatec, Germany). Microbial biomass C and N concentrations were calculated from these results using a kEC value of 0.45 (Wu et al., 1990) and a kEN
value of 0.54 (Brookes et al., 1985) and are presented in percent of 1 g of dry soil. The soil water
content was determined in another 10 g of soil, which was dried at 105°C. These soil samples and
the plant samples were ground with a ball mill before analysis. The C and N concentration in
shoots, roots, and soil was measured with a Euro EA C/N analyser (EuroVector, Italy).
A Thermo Finnigan MAT Delta plus Advantage isotope ratio mass spectrometer (IRMS) was
coupled to this C/N analyser to measure δ13C values in shoots, roots, and soil. Since only solid samples could be analysed by the IRMS unit, the CO2 and microbial biomass samples had to be specifically prepared. Any CO2 trapped as sodium carbonate (Na2CO3) in 5 ml of NaOH was precipitated
with 5 ml of 0.5 M strontium chloride (SrCl2) aqueous solution. To prevent fractionation in this
step, carbonate was completely precipitated to a maximum of 2.6 · 10–5% of the total CO2-C absorbed by NaOH remaining in the solution. The maximum residue in the NaOH solution was calculated according to the SrCO3 solubility product. The NaOH solutions containing the SrCO3
precipitants were then centrifuged three times at 3000 rev min–1 for 10 min and washed in between
with deionised and degassed water to remove NaOH and to reach a pH of 7. Keeping the tubes
opened for washing as short as possible prevented contamination by atmospheric CO2 during sample preparation. After washing, the remaining water was removed from the vials and the SrCO3 was
dried at 105°C. The SrCO3 was analysed on the IRMS for δ13C values. For the microbial biomass,
an aliquot of the K2SO4 samples was pipetted directly into tin capsules and dried at 60°C prior to
IRMS analyses. Drying of K2SO4 extracts in tin capsules prevented volatilisation of unstable compounds and additional 13C fractionation, which is typical for freeze-drying.
Calculations
A mass balance equation was used to determine the δ13C value of microbial biomass (δ13CMO):
46
3 Three-source partitioning of CO2 efflux from soil planted with maize
δ C MO =
13
δ 13C fum ⋅ C fum −δ 13C extr ⋅ C extr
C fum − C extr
(3-1),
where δ13Cfum and δ13Cextr are the δ13C values of the fumigated and extracted samples, respectively,
and Cfum and Cextr are the amounts of C in the fumigated and extracted K2SO4 samples, respectively.
In the beginning of every CO2 trapping there was a small volume of atmospheric CO2 in the
closed system, especially in soil pore space and in the trapping tube above the NaOH solution. We
considered this atmospheric CO2 from the measured δ13C value by a mass balance equation:
δ 13C corrected =
δ 13C total ⋅ C total −δ 13C air ⋅ C air
C total − C air
(3-2),
where δ13Ccorrected is the δ13C value of soil air without atmospheric air, δ13Ctotal is the measured δ13C
value of CO2, δ13Cair is the δ13C value of ambient air (–7.8‰, see Boutton (1991)), Ctotal is the
amount of CO2-C trapped in NaOH, and Cair is the amount of C in the soil pore space and the trapping tube in our closed system (0.024 mg C) calculated from a CO2 concentration of 345 mg kg–1
(Boutton, 1991) and the volume of air in the system.
After calculating the δ13C of microbial biomass (Eq. (3-1)) and the corrected δ13C of total CO2
efflux (Eq. (3-2)), it was possible to calculate below-ground CO2 partitioning. The development of
the equations used to calculate below-ground CO2 partitioning is presented in detail by Kuzyakov
(2004). The equations for SOMD and RMR are:
δ CO − δ 4Rhiz
2
SOMD =
(3-3)
δ 3SOM − δ 4Rhiz
(δ MO − δ3SOM ) ⋅ (δ CO2 − δ4Rhiz )
RMR = Rhiz
(δ4 − δ3SOM ) ⋅ (δ MO − δ4Rhiz )
(3-4),
where δCO2 is the δ13C value of the total CO2 efflux from planted soil, δ4Rhiz is the δ13C value of C4
plant roots, δ3SOM is the δ13C value of SOM from unplanted soil, and δMO is the δ13C value of microorganisms from planted soil. Having calculated these two contributions to the below-ground CO2
efflux, the remaining part would be RR:
RR = 1 − SOMD − RMR
(3-5).
A calculated δ13C value was used to determine the influence of active and inactive microbial
biomass fractions on δ13C of total microbial biomass. This δ13C value (δ13Ctotal) was calculated by a
mass balance equation using δ13C values of maize roots for active (δ13Cactive) and δ13C values of
SOM from unplanted soil for inactive (δ13Cinactive) portions of microbial biomass:
δ 13C total =
δ 13C active ⋅ C active +δ 13C inactive ⋅ C inactive
C total
(3-6),
3 Three-source partitioning of CO2 efflux from soil planted with maize
47
where Cactive, Cinactive, and Ctotal are amounts of C in active, inactive, and total microbial biomass
fractions, respectively. Ctotal was considered as 100%, Cactive was adjusted to match measured results
of below-ground CO2 partitioning (see in the results section), and Cinactive was Ctotal – Cactive.
Standard deviations (SD) were calculated as a variability parameter for all our results. We used a
one-way analysis of variance to identify differences between δ13C values of various below-ground
CO2 sources. The effect of 13C fractionation by microbial respiration on below-ground CO2 partitioning results was examined by a sensitivity analysis according to Kuzyakov (2005). δ13C of microbial CO2 was increased stepwise in this sensitivity analysis from 1‰ to 5‰ compared to
microbial biomass.
Results
C and N concentrations, C/N ratio, and cumulative CO2 efflux from soil
The C concentration in plant parts was constant during the entire experiment and averaged about 43
and 33% for shoots and roots, respectively (Table 3-1). The low C concentration in roots can be
explained by mineral soil particles remaining on roots after washing. Between days 16 and 40, the
total N concentration in the shoots decreased by 2.1% (Table 3-1). The shoots’ N concentrations
were about twice as high as in the roots. The N concentrations in both shoots and roots were expected to decrease because the plants grew and the amount of fertilization was held constant but not
increased. Consequently, on day 40, the C/N ratio increased to 30 in the shoots and 50 in the roots
(Table 3-1). C and N concentrations in the soil (Table 3-1) remained constant at 1.4% and 0.2%,
respectively. The soil C/N ratio was 9 on all sampling days. The C concentration in the microbial
biomass was only slightly increased on day 16, then remaining at about 0.022% of soil dry matter
on the following dates (Table 3-1). The N concentration in microbial biomass was also stable during the whole experiment. The C/N ratio of the microbial biomass was 2 units higher compared to
the bulk soil.
The cumulative CO2 efflux from the planted soil increased linearly by 10.7 mg C d–1 (Fig. 3-2).
In contrast, the control pots without plants showed a reduced rate of increase (2.7 mg C d–1). As a
first approximation of separate rhizosphere respiration and SOM decomposition, the latter curve
could be considered as CO2 derived from decomposition of SOM (up to 34% of total CO2 efflux
from planted soil). The difference between the two curves would then be rhizosphere respiration,
which amounted up to 66% of total CO2 efflux from planted soil. This difference approach between
planted and unplanted soil neglects interactions between enhanced microbial activity by rhizodeposition and the decomposition of SOM. Thus, it is only a rough estimate of C flows in the
rhizosphere.
48
3 Three-source partitioning of CO2 efflux from soil planted with maize
Table 3-1: Carbon and nitrogen concentrations and C/N ratios of shoots, roots, soil, and microbial biomass
on five sampling dates of maize grown on C3 soil (means ± SD, n = 4), based on plant part or soil dry matter.
Days of maize
growth
C
N
[% of dry matter] [% of dry matter]
C/N
shoots
16
22
28
34
40
42.0
42.5
42.6
41.4
45.0
±
±
±
±
±
2.3
1.1
2.0
2.5
1.7
3.6
2.7
2.1
1.7
1.5
±
±
±
±
±
0.4
0.3
0.1
0.2
0.1
11.6
15.9
20.3
24.0
30.0
±
±
±
±
±
0.6
1.4
0.5
2.4
2.0
roots
16
22
28
34
40
33.1
32.9
31.8
32.5
32.8
±
±
±
±
±
2.4
2.8
1.2
1.6
2.0
1.8
1.1
0.9
0.8
0.7
±
±
±
±
±
0.2
0.1
0.0
0.0
0.1
18.7
30.0
35.2
41.8
50.2
±
±
±
±
±
1.1
1.3
2.1
1.8
6.0
soil
16
22
28
34
40
1.5
1.5
1.4
1.5
1.4
±
±
±
±
±
0.1
0.0
0.2
0.0
0.1
0.2
0.2
0.2
0.2
0.2
±
±
±
±
±
0.0
0.0
0.0
0.0
0.0
9.0
9.0
9.1
9.1
9.0
±
±
±
±
±
0.4
0.1
0.8
0.3
0.2
microbial
biomass
16
22
28
34
40
0.031
0.020
0.024
0.021
0.021
±
±
±
±
±
0.003
0.003
0.002
0.003
0.001
0.002
0.002
0.002
0.002
0.002
±
±
±
±
±
0.001
0.000
0.000
0.000
0.000
11.3
10.7
11.2
11.2
11.1
±
±
±
±
±
0.9
0.1
0.7
1.2
2.0
Fig. 3-2: Cumulative CO2 efflux from C3 soil with maize (¡) and without plants ({); error bars show standard deviation (1 ≤ n ≤ 20, dependent on sampling date).
3 Three-source partitioning of CO2 efflux from soil planted with maize
49
δ13C values and CO2 efflux partitioning
Between days 16 and 40, the δ13C of maize roots slightly decreased, averaging –15.8‰ (Fig. 3-3a).
The δ13C of the total CO2 efflux from planted soil (–17.0‰) was significantly more negative (P <
0.05), by 1‰ compared to the roots. Nevertheless, δ13C values of roots and CO2 were very close.
This similarity indicates a high contribution of RR to the total CO2 efflux from the soil. The δ13C
values of CO2 presented in Fig. 3-3a were corrected by Eq. (3-2) for small amounts of air CO2 remaining in the soil pores and in the trapping tube. This correction made the δ13C values of belowground CO2 slightly more negative compared to uncorrected data, but this difference was less than
0.02‰. The δ13C of SOM was constant and amounted to –26.8‰. Until day 40, the δ13C of microbial biomass increased from –24.6‰ to –22.5‰; the mean value was –23.7‰, which was significantly more positive than the δ13C of SOM (P < 0.001).
The δ13C of SOM in unplanted soil (–27.0‰) was the same as in planted soil (Fig. 3-3b). In the
total CO2 efflux of unplanted soil, the mean δ13C between days 22 and 40 was –21.8‰. The mean
δ13C of microbial biomass between days 22 and 40 was intermediate between these two values
(–23.8‰). Consequently, there was a 13C fractionation of about 3.2‰ between organic matter in
unplanted soil and microbial biomass (P < 0.001), and of 2.0‰ between microbial biomass and
microbially respired CO2 (P < 0.05). The fractionation between SOM and microbial CO2 was 5.2‰
(P < 0.001).
We calculated contributions of RR, RMR, and SOMD to total CO2 efflux from the δ13C values in
Fig. 3-3 using Eqs. (3-3) to (3-5) (Fig. 3-4), which are based on the approach of Kuzyakov (2004).
The contributions of RR to total CO2 efflux were very dominant, with maxima of 91% on days 34
and 40. RMR was maximally only 9% and SOMD doubled this value at maximum.
Fig. 3-3: δ13C values of carbon pools in (a) maize grown for 40 days on a C3 soil and (b) C3 soil without
plants. Carbon pools are maize roots (…), soil organic matter (‘), total CO2 efflux (U), and microbial biomass (¯); error bars in (a) show standard deviation (n = 4); no error bars in (b) (n = 1).
50
3 Three-source partitioning of CO2 efflux from soil planted with maize
Fig. 3-4: Contributions of root respiration (no shading), rhizomicrobial respiration (hatched shading), and
SOM decomposition (dotted shading) to total CO2 efflux from a C3 soil planted with maize; error bars show
standard deviation (n = 4).
The portions of RR and RMR in rhizosphere respiration reported in other studies were about
50% each. In our experiment, there was a strong shift towards RR. Potential reasons for this shift
are (1) the above-mentioned difference in δ13C between microbial biomass and microbial CO2 and
(2) the discrepancy between the small active fraction of microbial biomass that feeds on rhizodeposits and the large fraction of microbially derived CO2 from active microbial biomass. Both reasons
are important, because we used δ13C from microbial biomass to calculate microbially derived CO2
assuming no fractionation between microbial biomass and microbial CO2 (see assumption 2). The
former case would have yielded underestimated contributions of microbial and rhizomicrobial CO2
to total CO2 efflux due to more negative δ13C values of microbial biomass compared to microbial
and rhizomicrobial CO2. In the latter case, δ13C of microbial biomass would have been mainly influenced by dormant microorganisms, which had fed formerly on SOM with C3 signature, leading
to a δ13C value close to C3 soil. However, the δ13C of rhizomicrobially respired CO2 would have
been mainly controlled by active microorganisms in the rhizosphere, which fed on rhizodeposits,
leading to a δ13C value close to that of C4 plants. These influences on the contributions of RR,
RMR, and SOMD will be presented in the following two sections.
Sensitivity analysis of changing 13C fractionation on below-ground CO2 partitioning
A sensitivity analysis was conducted to determine the effect of 13C fractionation by microbial respiration on a mean of the CO2 partitioning results from days 34 and 40. The δ13C of microbial CO2
was increased stepwise from 1‰ to 5‰ (Fig. 3-5). A maximum 13C fractionation of 5‰ compared
to δ13C of microbial biomass increased RMR up to 32% and decreased RR down to 62% of total
CO2 efflux. The contribution of microbial SOM decomposition was not affected by 13C fractiona-
3 Three-source partitioning of CO2 efflux from soil planted with maize
51
tion during microbial respiration. To determine the latter, δ13C values of CO2 efflux and microbial
biomass from unplanted soil were monitored from day 10 to 40 (Fig. 3-3b). The difference between
these δ13C values showed a mean 13C fractionation of 2.0‰ with a 13C enrichment in the CO2. Considering this 13C fractionation in mass balance Eq. (3-4), the contributions of RR, RMR, and SOMD
to total CO2 efflux amounted to 87, 7, and 6%, respectively.
Fig. 3-5: Sensitivity analysis of 13C fractionation between microbial biomass (δ13C = –22.7‰) and microbial
CO2 (δ13C = –22.7‰ + 0 to 5‰) on contributions of root respiration (no shading), rhizomicrobial respiration
(hatched shading), and SOM decomposition (dotted shading) to total CO2 efflux from a C3 soil planted with
maize; mean CO2 efflux contributions are built from days 34 and 40, error bars show standard deviation (n =
2).
Effect of active microbial biomass on below-ground CO2 partitioning
Isotopic 13C fractionations of 2.0‰ between microbial biomass and microbial CO2 and of 5.2‰
between SOM and SOM-derived CO2 were accounted for in this approach. Using these fractionations and the δ13C values from Fig. 3-3, we calculated the partitioning of CO2 efflux from soil for a
mean of the last two sampling dates (left column in Fig. 3-6). In order to simulate the influence of
active and inactive fractions of the microbial biomass on CO2 partitioning, we used calculated δ13C
values for the microbial biomass that considered both fractions (Eq. (3-6)). Percentages of these
fractions in Eq. (3-6) were adjusted to match the CO2 partitioning results obtained in this study
(middle column in Fig. 3-6) and literature results (right column in Fig. 3-6). Values of δ13C for roots
(–16.2‰) and for SOM (–26.9‰) were used to represent δ13C values for active and inactive microbial biomass fractions, respectively.
An active portion, which feeds on maize rhizodeposits, of about 37% of total microbial biomass
(middle column in Fig. 3-6) was determined to reflect the results observed in this study (left column
52
3 Three-source partitioning of CO2 efflux from soil planted with maize
in Fig. 3-6). A hypothetical active portion of 71% of total microbial biomass (right column in Fig.
3-6), however, would have been necessary to yield a 50% contribution each for RR and RMR related to total rhizosphere respiration as reported in various studies (Cheng et al., 1993; Kuzyakov et
al., 2001).
Fig. 3-6: Influence of active portion of microbial biomass on below-ground CO2 partitioning. In the middle
column the active portion was adjusted to 37% of total microbial biomass to achieve calculated CO2 partitioning results of this present study for a mean of days 34 and 40 (left column). In the right column the active
portion of microbial biomass was adjusted to 71% of total microbial biomass to achieve CO2 partitioning
results of literature studies (see introduction). Patterns are: contributions of root respiration (no shading),
rhizomicrobial respiration (hatched shading), and SOM decomposition (dotted shading) to total CO2 efflux
from a C3 soil planted with maize; error bars show standard deviation (n = 2).
Discussion
Evaluation of the natural 13C labelling technique for below-ground CO2 partitioning
On the last two sampling days (34 and 40 after germination), the δ13C values and the partitioning of
the below-ground CO2 efflux showed that the plant–soil systems had stabilised (Fig. 3-3a and 3-4).
Root respiration was strongly overestimated by the examined approach of Kuzyakov (2004). Rhizomicrobial respiration and SOM decomposition were both remarkably underestimated. Two indications point to the incorrect estimate of CO2 partitioning found using the 13C labelling technique:
1) The results of the cumulative CO2 efflux of planted and unplanted soil show much lower
portions of rhizosphere respiration (66%) and higher portions of SOM decomposition (34%),
which fit very well with literature results (Ekblad and Högberg, 2001; Kuzyakov and Cheng,
2001).
3 Three-source partitioning of CO2 efflux from soil planted with maize
53
2) From the literature reviewed in the introduction, we have calculated that RR and RMR each
contribute equally (50%) to rhizosphere respiration.
Based on these two considerations the results obtained by the natural 13C labelling technique in this
study cannot be accepted.
Verification of assumptions on 13C fractionation
The below-ground CO2 partitioning results change slightly if we consider the two assumptions from
the introduction. The first assumption – equal δ13C values of roots and rhizosphere respiration – has
been used in most rhizosphere CO2 studies to date (Cerling et al., 1991; Amundson et al., 1998; Fu
and Cheng, 2002). The study of Cheng (1996), in which winter wheat was grown on C-free vermiculite and on a vermiculite-sand-mixture, proves this assumption. Even if fractionation occurs in
this process, it should be very small, and root-respired CO2 should only be about 0.7‰ depleted in
13
C compared to roots (Werth and Kuzyakov, 2005). Hence, the first assumption has to be accepted.
The second assumption – equal δ13C values of microbial biomass and microbial CO2 – was
checked in the literature: we found not only 13C fractionations between microbial biomass and CO2,
but also between microbial biomass and SOM and between SOM and CO2. The results vary
strongly for the first fractionation between microbial biomass and CO2. According to Šantrůčková
et al. (2000a), δ13C values of CO2 respired from 21 Australian soils with C3 and C4 vegetation were
depleted on average by +2.2‰ compared to microbial biomass. For individual soils, the δ13C difference between microbial biomass and respired CO2 varied between +0.1‰ and +5.7‰. Our results,
however, showed a 13C enrichment of CO2 by 2.0‰ compared to microbial biomass (Fig. 3-3b).
This contradiction needs to be discussed relative to the second and third fractionation.
For the second fractionation between microbial biomass and SOM, we observed on average
about 3.2‰ higher δ13C values in the microbial biomass compared to SOM in unplanted soil samples. Results of recent studies confirm this fractionation (Ryan et al., 1995; Šantrůčková et al.,
2000a; Potthoff et al., 2003). Isotope discrimination during biosynthesis of new microbial biomass
and the heavier isotopic composition of organic compounds preferentially used by soil microorganisms explain this 13C enrichment in microbial biomass (Potthoff et al., 2003).
The third fractionation between SOM as the substrate and microbial CO2 as the product is the
sum of the first and the second fractionation. Usually, CO2 from microbial respiration is 13Cdepleted compared to the feeding substrate (Blair et al., 1985; Mary et al., 1992; Potthoff et al.,
2003). In a study of Šantrůčková et al. (2000b) the difference between δ13C of SOM and that of
respired CO2 varied between +0.5‰ and –1.7‰. Formanek and Ambus (2004) reported a 13C enrichment of respired CO2 compared to SOM between 3.6‰ and 5‰. These results imply a 13C enrichment of CO2 compared to the substrate in most cases. Such an enrichment agrees with our
results from unplanted soil (Fig. 3-3b) and indicates that only a 13C-enriched fraction of the total
organic C was used in these mineralisation processes. This isotope effect associated with the selec-
54
3 Three-source partitioning of CO2 efflux from soil planted with maize
tive use of organic compounds was more pronounced than the 13C depletion effect of metabolism
itself (Šantrůčková et al., 2000b). The use of this 13C-enriched SOM fraction leads to a more rapid
loss of 13C than 12C during decomposition and therefore depletes the 13C in the remaining material
(Benner et al., 1987; Ågren et al., 1996). These results led us to use a 5.2‰ fractionation between
SOM and CO2 in considering effects of active microbial biomass on below-ground CO2 partitioning.
Fractionations in the CO2 sampling and sample preparation can be excluded because we eliminated the influence of atmospheric CO2 by calculating corrected δ13C values according to Eq. (3-2)
and because we completely precipitated the CO2 by SrCl2 solution. Consequently, our second assumption cannot be accepted because 13C isotopic fractionations between microbial biomass and
microbial CO2, and between SOM and CO2 from its decomposition, remain the two most important
sources of error in below-ground CO2 partitioning. These two fractionations have to be measured
under experimental conditions.
The enormous impact of 13C fractionation between microbial biomass and CO2 on RMR is evident in the 3 to 5‰ fractionation range in our sensitivity analysis (Fig. 3-5). Nevertheless, even
with 5‰ fractionation, the results of former studies (50% RR and RMR each in relation to
rhizosphere respiration) and the results of our cumulative CO2 efflux from planted and unplanted
soil (66% rhizosphere respiration, 34% SOM decomposition) could not be achieved by the tested
isotopic approach. Due to a shift in the δ13C value of microbial respiration with an increase of fractionation towards the δ13C value of the maize roots, the impact of this fractionation on SOMD was
visible only at the second decimal place.
Influence of active microbial biomass on below-ground CO2 partitioning
Since only a minor part of microbial biomass is metabolically active in soil (Stenström et al., 2001),
we examined the effect of active microbial biomass on below-ground CO2 partitioning. In both
cases – the one matching our measured results and the other one matching literature results – the
active microbial biomass fraction (37 to 71% of total microbial biomass) is rather high (Fig. 3-6).
Especially for this short 40-day period, other studies showed much lower maximum values of 6 to
23% (Bruulsema and Duxbury, 1996; Qian and Doran, 1996; Rochette et al., 1999). Thus, 37%
active microbial biomass gives a very high estimate to explain our results, and 71% is only a theoretical value to approximate our data to literature data. Thus, a very high active microbial fraction
would be necessary to match the literature results (50% RR and 50 % RMR contribution to
rhizosphere respiration) – an absolutely unrealistic value in a real ecosystem. Our calculations assumed that active microbial biomass feeds solely on rhizodeposits. Clearly, some microorganisms
also feed on SOM. Our calculated active fraction would therefore be slightly bigger when including
the latter microorganisms. The inactive microbial biomass fraction would be correspondingly
smaller.
3 Three-source partitioning of CO2 efflux from soil planted with maize
55
Besides the large contribution of inactive SOM-feeding organisms to the microbial biomass,
Bruulsema and Duxbury (1996) assumed that the chloroform fumigation method solubilises a substantial fraction of less active non-microbial soil organic C. Consequently, the natural 13C labelling
method fails due to (1) a low active microbial biomass fraction and/or (2) chloroform-soluble
nonliving organic material.
Conclusions
The isotopic mass balance from soil planted with maize was insufficient to accurately partition total
CO2 efflux into three CO2 sources: RR, RMR, and SOMD. The method strongly overestimated RR
and underestimated RMR and SOMD. The main problem of the approach was the strong discrepancy between δ13C values of CO2 respired by microbial biomass and of microbial biomass itself,
indicating that only a small portion of active microorganisms utilized maize rhizodeposits. Besides
this discrepancy, isotopic fractionation during SOM decomposition and microbial biomass respiration should be estimated in a separate experiment with unplanted soil; the results should be considered in all calculations.
Mathematically changing the portion of active microbial biomass showed that this microbial
biomass is mainly responsible for altered RR and RMR portions. To attain the partitioning results of
other studies, the portion of active microbial biomass would have to be at least 71%. We conclude
that the three-sources CO2 partitioning approach using a natural 13C labelling technique failed and
do not recommend its use in future studies.
Acknowledgements
The German Research Foundation (DFG) supported this study. The authors would also like to thank
Dr. W. Armbruster and E. Dachtler for the IRMS analyses.
References
Ågren, G.I., Bosatta, E., Balesdent, J., 1996. Isotope discrimination during decomposition of organic matter: a theoretical analysis. Soil Science Society of America Journal 60, 1121-1126.
Amundson, R., Stern, L., Baisden, T., Wang, Y., 1998. The isotopic composition of soil and soilrespired CO2. Geoderma 82, 83-114.
Balesdent, J., Mariotti, A., 1996. Measurement of soil organic matter turnover using 13C natural
abundance. In: Boutton, T.W., Yamasaki, S.I. (Eds.), Mass Spectrometry of Soils, Marcel
Dekker, New York, USA, pp. 83-111.
Benner, M.H., Hendrix, P.F., Coleman, D.C., 1987. Depletion of 13C in lignin and its implications for stable carbon isotope studies. Nature 329.
56
3 Three-source partitioning of CO2 efflux from soil planted with maize
Blair, N., Leu, A., Munoz, E., Olsen, J., Kwong, E., Des Marais, D., 1985. Carbon isotope fractionation in heterotrophic microbial metabolism. Applied and Environmental Microbiology
50, 996-1001.
Boutton, T.W., 1991. Stable carbon isotope ratios of natural materials: II. atmospheric, terrestrial,
marine, and freshwater environments. In: Coleman, D.C., Fry, B. (Eds.), Carbon Isotope
Techniques. Isotopic Techniques in Plant, Soil, and Aquatic Biology, Academic Press, Inc.,
San Diego, pp. 173-185.
Brookes, P.C., Landman, A., Pruden, G., Jenkinson, D.S., 1985. Chloroform fumigation and the
release of soil nitrogen: a rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biology & Biochemistry 17, 837-842.
Bruulsema, T.W., Duxbury, J.M., 1996. Simultaneous measurement of soil microbial nitrogen,
carbon, and carbon isotope ratio. Soil Science Society of America Journal 60, 1787-1791.
Cerling, T.E., Solomon, D.K., Quade, J., Bowman, J.R., 1991. On the isotopic composition of
carbon in soil carbon dioxide. Geochimica et Cosmochimica Acta 55, 3404-3405.
Cheng, W., 1996. Measurement of rhizosphere respiration and organic matter decomposition using
natural 13C. Plant and Soil 183, 263-268.
Cheng, W., Coleman, D.C., Carroll, C.R., Hoffman, C.A., 1993. In situ measurement of root
respiration and soluble C concentrations in the rhizosphere. Soil Biology & Biochemistry 25,
1189-1196.
Edwards, N.T., Harris, W.F., 1977. Carbon cycling in a mixed deciduous forest floor. Ecology 58,
431-437.
Ekblad, A., Högberg, P., 2001. Natural abundance of 13C in CO2 respired from forest soils reveals
speed of link between tree photosynthesis and root respiration. Oecologia 127, 305-308.
Formánek, P., Ambus, P., 2004. Assessing the use of δ13C natural abundance in separation of root
and microbial respiration in a Danish beech (Fagus sylvatica L.) forest. Rapid Communications in Mass Spectrometry 18, 1-6.
Fu, S., Cheng, W., 2002. Rhizosphere priming effects on the decomposition of soil organic matter
in C4 and C3 grassland soils. Plant and Soil 238, 289-294.
Högberg, P., Nordgren, A., Buchmann, N., Taylor, A.F.S., Ekblad, A., Högberg, M.N., Nyberg, G., Ottosson-Löfvenius, M., Read, D.J., 2001. Large-scale forest girdling shows that
current photosynthesis drives soil respiration. Nature 411, 789-792.
Johansson, G., 1992. Release of organic C from growing roots of meadow fescue (Festuca pratensis L.). Soil Biology & Biochemistry 24, 427-433.
Killham, K., Yeomans, C., 2001. Rhizosphere carbon flow measurement and implications: from
isotopes to reporter genes. Plant and Soil 232, 91-96.
Kuzyakov, Y., 2002. Separating microbial respiration of exudates from root respiration in nonsterile soils: a comparison of four methods. Soil Biology & Biochemistry 34, 1621-1631.
Kuzyakov, Y., 2004. Separation of root and rhizomicrobial respiration by natural 13C abundance:
theoretical approach, advantages, and difficulties. Eurasian Soil Science 37, S79-S84.
Kuzyakov, Y., 2005. Theoretical background for partitioning of root and rhizomicrobial respiration
by δ13C of microbial biomass. European Journal of Soil Biology 41, 1-9.
Kuzyakov, Y., Siniakina, S.V., 2001. A novel method for separating root-derived organic compounds from root respiration in non-sterilized soils. Journal of Plant Nutrition and Soil Science 164, 511-517.
Kuzyakov, Y., Cheng, W., 2001. Photosynthesis controls of rhizosphere respiration and organic
matter decomposition. Soil Biology & Biochemistry 33, 1915-1925.
3 Three-source partitioning of CO2 efflux from soil planted with maize
57
Kuzyakov, Y., Domanski, G., 2002. Model for rhizodeposition and CO2 efflux from planted soil
and its validation by 14C pulse labelling of ryegrass. Plant and Soil 239, 87-102.
Kuzyakov, Y., Larionova, A.A., 2005. Root and rhizomicrobial respiration: A review of approaches to estimate respiration by autotrophic and heterotrophic organisms in soil. Journal
of Plant Nutrition and Soil Science 168, 503-520.
Kuzyakov, Y., Kretzschmar, A., Stahr, K., 1999. Contribution of Lolium perenne rhizodeposition
to carbon turnover of pasture soil. Plant and Soil 213, 127-136.
Kuzyakov, Y., Ehrensberger, H., Stahr, K., 2001. Carbon partitioning and below-ground translocation by Lolium perenne. Soil Biology & Biochemistry 33, 61-74.
Mary, B., Mariotti, A., Morel, J.L., 1992. Use of 13C variations at natural abundance for studying
the biodegradation of root mucilage, roots and glucose in soil. Soil Biology & Biochemistry
24, 1065-1072.
Potthoff, M., Loftfield, N., Buegger, F., Wick, B., John, B., Jörgensen, R.G., Flessa, H., 2003.
The determination of δ13C in soil microbial biomass using fumigation-extraction. Soil Biology & Biochemistry 35, 947-954.
Qian, J.H., Doran, J.W., 1996. Available carbon released from crop roots during growth as determined by carbon-13 natural abundance. Soil Science Society of America Journal 60, 828831.
Rochette, P., Angers, D.A., Flanagan, L.B., 1999. Maize residue decomposition measurement
using soil surface carbon dioxide fluxes and natural abundance of carbon-13. Soil Science
Society of America Journal 63, 1385-1396.
Ryan, M.C., Aravena, R., Gillham, R.W., 1995. The use of 13C natural abundance to investigate
the turnover of the microbial biomass and active fractions of soil organic matter under two
tillage treatments. In: Lal, R., Kimble, J., Levine, E., Stewart, B.A. (Eds.), Soils and Global
Change, CRC Press, Boca Raton, pp. 351-360.
Šantrůčková, H., Bird, M.I., Lloyd, J., 2000a. Microbial processes and carbon-isotope fractionation in tropical and temperate grassland soils. Functional Ecology 14, 108-114.
Šantrůčková, H., Bird, M.I., Frouz, J., Šustr, V., Tajovský, K., 2000b. Natural abundance of 13C
in leaf litter as related to feeding activity of soil invertebrates and microbial mineralisation.
Soil Biology & Biochemistry 32, 1793-1797.
Stenström, J., Svensson, K., Johansson, M., 2001. Reversible transition between active and dormant microbial states in soil. FEMS Microbiology Ecology 36, 93-104.
Swinnen, J., 1994. Evaluation of the use of a model rhizodeposition technique to separate root and
microbial respiration in soil. Plant and Soil 165, 89-101.
Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring soil microbial biomass C. Soil Biology & Biochemistry 19, 703-707.
Werth, M., Kuzyakov, Y., 2005. Below-ground partitioning (14C) and isotopic fractionation (δ13C)
of carbon recently assimilated by maize. Isotopes in Environmental and Health Studies 41,
237-248.
Wu, J., Jörgensen, R.G., Pommerening, B., Chaussod, R., Brookes, P.C., 1990. Measurement of
soil microbial biomass-C by fumigation-extraction – an automated procedure. Soil Biology
& Biochemistry 22, 1167-1169.
Zibilske, L.M., 1994. Carbon Mineralization. In: Weaver, R.W., Angle, S., Bottomley, P., Bezdicek, D., Smith, S., Tabatabai, A., Wollum, A. (Eds.), Methods of Soil Analysis, Part 2,
Microbiological and Biochemical Properties. Soil Science Society of America Book Series,
No. 5, Soil Sci. Soc. Am., Inc., Madison, pp. 835-864.
4
Root-derived carbon in soil respiration and microbial biomass
determined by 14C and 13C
Martin Werth and Yakov Kuzyakov
Soil Biology & Biochemistry (2007), in press, doi:10.1016/j.soilbio.2007.09.022
includes three tables and five figures
with kind permission of Elsevier Science Ltd
60
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
Abstract
Two approaches to quantitatively estimating root-derived carbon in soil CO2 efflux and in microbial
biomass were compared under controlled conditions. In the 14C labelling approach, maize (Zea
mays) was pulse labelled and the tracer was chased in plant and soil compartments. Root-derived
carbon in CO2 efflux and in microbial biomass was estimated based on a linear relationship between
the plant shoots and the below-ground compartment. Since the maize plants were grown on C3 soil,
in a second approach the differences in 13C natural abundance between C3 and C4 plants were used
to calculate root-derived carbon in the CO2 efflux and in the microbial biomass. The root-derived
carbon in the total CO2 efflux was between 69 and 94% using the 14C labelling approach and between 86 and 94% in the natural 13C labelling approach. At a 13C fractionation measured to be 5.2‰
between soil organic matter (SOM) and CO2, the root-derived contribution to CO2 ranged from 70
to 88% and was much closer to the results of the 14C labelling approach. Root-derived contributions
to the microbial biomass carbon ranged from 2 to 9% using 14C labelling and from 16 to 36% using
natural 13C labelling. At a 3.2‰ 13C fractionation between SOM and microbial biomass, both labelling approaches yielded an equal contribution of root-derived C in the microbial biomass. Both approaches may therefore be used to partition CO2 efflux and to quantify the C sources of microbial
biomass. However, the assumed 13C fractionation strongly affects the contributions of individual C
sources.
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
61
Introduction
Carbon dioxide (CO2) efflux from soils is an important component of the global carbon (C) cycle
and related to global climatic change because increasing amounts of CO2 in the atmosphere promote
the greenhouse effect. Small changes in the turnover intensity of soil organic matter (SOM) could
significantly alter the CO2 concentration in the atmosphere: the amount of C in SOM approximately
doubles the amount of C in the atmosphere (Grace, 2004). These small variations in the decomposition intensity of SOM cannot be determined directly by measuring organic carbon contents because
changes in soil organic C are very small during short periods (e.g. 1-3% during a single vegetative
growth season). Alternatively, measuring CO2 efflux from soil is commonly used to investigate
short-term SOM turnover. This method is sensitive enough to detect small and actual changes, especially for recently altered ecosystems (Kuzyakov and Cheng, 2001). Most soils, however, are
covered with vegetation, which also contributes to the CO2 efflux from soil. Therefore, CO2 efflux
from planted soil consists of SOM- and root-derived CO2. The latter can be further subdivided into
root respiration and rhizomicrobial respiration of rhizodeposits (exudates, lysates, etc.). This separation is exceptionally difficult, since plant roots and rhizosphere microorganisms use the same carbon source, i.e. plant assimilates. It is much easier to separate CO2 from microbial decomposition of
soil organic matter and root-derived CO2, i.e. the sum of root respiration and respiration of
rhizosphere microorganisms consuming rhizodeposits. Since root-derived CO2 is not part of soil C
loss, partitioning the total CO2 efflux from soil is very important to identify individual sinks or
sources of CO2.
Carbon dioxide derived from soil organic matter decomposition and that derived from the roots
can be partitioned and quantified by isotopic labelling of plants with 13C or 14C isotopes and tracing
the label in root-derived CO2 (Warembourg and Paul, 1977; Andrews et al., 1999). Besides artificial labelling techniques, the difference in the natural abundances of 13C in C3 plants (–35‰ ≤ δ13C
≤ –20‰) and in C4 plants (–15‰ ≤ δ13C ≤ –7‰) can also be used as a natural carbon tracer (Cheng,
1996; Qian et al., 1997; Rochette and Flanagan, 1997; Ekblad and Högberg, 2001; Kuzyakov and
Cheng, 2001). The difference between the labelled fraction and the total CO2 efflux represents CO2
from SOM decomposition. Non-isotopic methods to separate root- from SOM-derived CO2, such as
a combination of trenching and excised-root methods, have also been used (Kelting et al., 1998;
Chen et al., 2006). The results vary strongly depending on plants, soils, and environmental and experimental conditions. By in situ 14C labelling of Canadian prairie grass, Warembourg and Paul
(1977) found low contributions (19%) of root-derived CO2 to the total CO2 efflux from soil. On the
other hand, under controlled conditions, Chen et al. (2006) reported very high contributions of rootderived CO2, with values of up to 99% in a ryegrass (Lolium perenne L.) rhizosphere. Various studies under controlled conditions have found results within this range (Robinson and Scrimgeour,
1995; Qian et al., 1997; Kuzyakov and Cheng, 2001; Lin et al., 2001), with an average contribution
62
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
of 59±23% root-derived CO2. The broad variability of these results indicates that there is urgent
need to find a reproducible standard method including a protocol for standardized soil preparation,
plant age and growing conditions, and analytical procedures.
The turnover of SOM and rhizodeposits is caused by the soil microbial biomass, which derives
its energy from oxidising soil organic C. Both the plant residue C and the rhizodeposits pass
through the soil microbial biomass at least once as they are transferred from one C pool to another
and finally mineralised to CO2 (Ryan and Aravena, 1994). In a system of C3-C4-vegetation change,
active microorganisms can be identified by high contributions of the C4-source to their δ13C signature, since this is an indicator of food uptake from recently assimilated C. Alternatively, active
rhizosphere microorganisms can be determined by the 14C tracer after labelling of plants followed
by a rhizodeposition of this tracer. Root-derived carbon – i.e. C4-derived carbon – ranges for instance from 9 to 52% in the soil microbial biomass after 110 days of maize growth (Liang et al.,
2002). Several other studies using 13C or 14C labelling techniques have found contributions of rootderived carbon within this range (Merckx et al., 1987; Ryan and Aravena, 1994; Angers et al.,
1995; Bruulsema and Duxbury, 1996; Qian and Doran, 1996; Rochette et al., 1999; Gregorich et
al., 2000).
Fractionations between the substrate, the microbial biomass, and the microbially respired CO2
have not always been considered in earlier studies or it was assumed that the fractionation is not
significantly different from zero (Cheng, 1996; Ekblad and Högberg, 2000; Ekblad et al., 2002).
However, control treatments without plants allow these fractionations to be determined. Several
studies have considered 13C fractionations between the substrate, the microbial biomass, and the
CO2 (Mary et al., 1992; Schweizer et al., 1999; Šantrůčková et al., 2000; Fernandez and Cadisch,
2003; Kristiansen et al., 2004). In order to identify the impact of isotopic fractionation on rootderived carbon contributions, we used the natural 13C labelling technique with and without consideration of 13C fractionation.
Determining root-derived contributions to below-ground carbon pools using the 14C pulse labelling technique and the natural 13C labelling technique has often led to different, sometimes contrasting results. This is because both methods are based on different assumptions, their sensitivity
strongly differs, and the distributions of the tracer could vary. The 14C pulse labelling technique
allows the distribution of recently assimilated C at specific plant development stages to be determined, but the partitioning of the tracer into plant and soil pools has to be completed on the sampling date. The distribution of plant-derived carbon to below-ground pools can only be determined
for the whole growth period by repeated labelling pulses. In contrast, natural 13C labelling is equivalent to a continuous labelling approach, which does not focus on recently assimilated carbon but on
the total plant-derived carbon in plant and soil pools, i.e. sampling can be done at any time. On
short time scales, however, both methods should produce similar results. It is unclear whether differences between the two methods reflect differences in plants, soils, experimental conditions etc. or
whether they are methodological artefacts. This calls for applying both methods under exactly the
same experimental conditions, preferably in the same experiment.
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
63
The objective of this study was to determine the contributions of maize-root-derived carbon to
the CO2 efflux from soil and to the soil microbial biomass. Two approaches were compared: (a) the
14
C pulse labelling approach and (b) the natural 13C labelling technique. In the former, maize plants
were artificially labelled with 14CO2 and the tracer was chased in plant and soil pools. The amount
of total root-derived carbon in CO2 or microbial biomass was then calculated with a linear function
according to Kuzyakov et al. (1999). In the natural 13C labelling technique (Balesdent and Mariotti,
1996), 13C natural abundance was used by growing maize as a C4 plant on a soil developed solely
under C3 vegetation (‘C3 soil’). Hence, four specific δ13C values were used in mass balances to determine the contributions of root-derived carbon to CO2 efflux and microbial biomass.
Materials and methods
Experimental set-up
Maize plants (Zea mays L.) were grown under controlled laboratory conditions in 20 pots filled
with a loamy Haplic Luvisol from loess with C3 vegetation history (Lolium perenne L.), collected
from the University of Hohenheim’s research farm ‘Heidfeldhof’ in Stuttgart, Germany. The maize
seeds (cv. Tassilo) were germinated on wet filter paper. One day after germination the seedlings
were transferred to 250 ml polycarbonate filtration devices (SM16510/11, Sartorius, Göttingen,
Germany) filled with 400 g of the C3 soil (pH(CaCl2) = 6.0), one plant per container (Fig. 4-1). A
control treatment with one unplanted pot per sampling date was established, which was treated exactly in the same way like the planted treatment. One day before the start of CO2 trapping, the holes
in the pots around the plant shoots were sealed with a 1-cm-thick silicone rubber layer (TACOSIL
145, Thauer & Co., Dresden, Germany) between roots and shoots, and the seal was tested for air
leaks. Trapping of CO2 from soil air started on day 9 after germination in a closed system for each
plant (or control treatment). Air was pumped through every single pot from bottom to top by a
membrane pump (Type 113, Rietschle Thomas, Memmingen, Germany, pumping rate 100 ml
min–1), which was connected to the pot by a polyvinyl chloride (PVC) tube (Fig. 4-1). Another PVC
tube was connected to the top outlet of the filter device and to a CO2 trapping tube filled with 20 ml
1 M sodium hydroxide (NaOH) solution. The output of the trapping tube was connected to the input
of the membrane pump. Therefore, the air containing CO2 evolved from soil respiration circulated
in a closed system. Firstly, the air was pumped through the pot, with any CO2 from total soil respiration being trapped in NaOH solution. Secondly, the remaining CO2-free air coming from the
NaOH trapping tube was pumped back through the pot. Thus, the air cycling was closed and was
done continuously by the membrane pump. This completely prevented CO2 losses and contamination with air CO2.
The soil moisture was maintained at about 25% of the gravimetrical water content throughout the
experiment by controlling the pots’ weights after the first water addition. On days 9, 15, 21, 27, and
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
64
33 after germination, a full fertilizer (5 kg nitrate-N ha–1, 0.4 kg monophosphate-P ha–1, 10 kg K+
ha–1; see Werth and Kuzyakov (2005) for further details) was added with the water to the soil from
one to five times depending on the date of sampling of the pots.
Silicone
rubber
NaOH
Soil
Rubber
rings
Filter
support
Membrane
pump
14
Fig. 4-1: Experimental set-up for trapping of below-ground CO2 in NaOH solution (redrawn from
Werth et al., 2006). White arrows show airflow.
C pulse labelling
On day 9 after germination, the 20 maize plants were labelled for the first time. All sealed pots with
plants were placed into a Plexiglas chamber (0.5 x 0.5 x 0.6 m³) for the labelling procedure described in detail by Cheng et al. (1993). Briefly, the chamber was connected by tubing with a flask
containing 2.0 ml 1 mM Na214CO3 solution to which 5 ml 9 M H2SO4 was added to produce 14CO2.
The plants were labelled during 2.5 h in the 14CO2 atmosphere. Usually, about 30 min of labelling
time are required for C4 plants to reach the CO2 compensation point (Kuzyakov and Cheng, 2004).
A longer time period was used in our experiment to increase the 14C incorporation into plant bio-
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
65
mass. Before opening the labelling chamber, the chamber air was pumped through 1 M NaOH solution to remove unassimilated 14CO2. Activities of unassimilated 14CO2 and of the 14C residue in the
Na214CO3 source were subtracted from the total 14C present in the flask prior to labelling in order to
calculate the total 14C input activity. The latter was divided by the number of plants in the labelling
chamber, yielding an input activity of 246.7 kBq per plant. After labelling, the chamber was opened
and the trapping of CO2 evolved by root respiration was started. The same labelling procedure was
repeated on days 15, 21, 27, and 33 with a total of 16, 12, 8, and 4 plants in the chamber, respectively. The 14C input activity was adjusted by the reduced numbers of plants in the labelling chamber (0.1 ml 1 mM Na214CO3 solution per plant).
Sampling and analyses
One week after labelling, soil and plants were destructively sampled in four replicates (i.e. one replicate for the control treatment) on days 16, 22, 28, 34, and 40 after germination. At harvest, each
shoot was cut at the base, the lid of the pot was opened and each root–soil column pulled out of the
pot. Bulk soil was sampled by cutting a small wedge into the soil column from the edge towards the
centre. We then loosened the soil column from the edge and discarded the soil falling down. The
soil still adhering to the roots was collected as the inner rhizosphere fraction and was used later on
for microbial biomass δ13C analyses. The moist soil samples were immediately frozen until preparation for microbial biomass. The roots were carefully washed with deionised water to remove soil
particles. Shoots and roots were dried at 40°C. Carbon dioxide trapped in NaOH was sampled on
the harvest days and additionally once to twice between two harvest days.
To estimate total CO2 efflux, the CO2 trapped in NaOH solution was precipitated with 0.5 M barium chloride (BaCl2) solution and then the NaOH was titrated with 0.2 M hydrochloric acid (HCl)
against phenolphthalein indicator (Zibilske, 1994). Soil microbial biomass was determined by the
chloroform fumigation extraction method (modified after Vance et al., 1987): roots were removed
from the unfrozen soil by handpicking, and 10 g of soil were extracted with 40 ml of 0.05 M potassium sulphate (K2SO4) solution. Another 10 g of soil were firstly fumigated with chloroform for 24
h and then extracted in the same way. The K2SO4 and soil mixtures were shaken for 1 h at 200 rev
min–1, centrifuged at 3000 rev min–1 for 10 min, and then filtrated through a ceramic vacuum filter.
The extracts were frozen until analyses for total carbon concentrations were done with a Dimatoc100 TOC/TIC analyser (Dimatec, Essen, Germany). The microbial biomass C concentration was
calculated from these results using a kEC value of 0.45 (Wu et al., 1990) and is presented in percent
of dry soil. The soil water content was determined in another 10 g of soil, which was dried at
105°C. These soil samples and the plant samples were ground with a ball mill before analysis. The
C concentration in shoots, roots, and soil was measured with a Euro EA C/N analyser (EuroVector,
Milan, Italy).
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
66
The 14C activity of 14CO2 trapped in NaOH solution was measured in 2 ml aliquots added to 4 ml
scintillation cocktail Rotiszint Eco Plus (Carl Roth, Karlsruhe, Germany) after decay of chemiluminescence. 14C activity was measured using a Wallac 1411 Liquid Scintillation Counter (Wallac Oy,
Turku, Finland). The 14C counting efficiency was about 85% and the 14C activity measurement error
did not exceed 2%. The absolute 14C activity was standardised by addition of NaOH solution as
quencher to the scintillation cocktail and using the spectrum of an external standard (SQP(E)
method). 14C in solid samples (dried shoots, roots, and soil) was measured on the liquid scintillation
counter after combustion of 200 mg of plant samples or 1 g of soil samples within an oxidizer unit
(Model 307, Canberra Packard Ltd., Meriden, USA), absorption of the 14C in Carbo-Sorb E (Perkin
Elmer, Inc., Boston, USA), and addition of the scintillation cocktail Permafluor E+ (Perkin Elmer,
Inc.).
A Thermo Finnigan MAT Delta plus Advantage isotope ratio mass spectrometer (IRMS from
Thermo Electron Corporation, Waltham, USA) was coupled to the C/N analyser to measure δ13C
values in shoots, roots, and soil. Since only solid samples could be analysed by the IRMS unit, the
CO2 and microbial biomass samples had to be specifically prepared. Any CO2 trapped as sodium
carbonate (Na2CO3) in 5 ml of NaOH was precipitated with 5 ml of 0.5 M strontium chloride
(SrCl2) aqueous solution. To prevent fractionation in this step, carbonate was completely precipitated to a maximum of 2.6 · 10–5% of the total CO2-C absorbed by NaOH remaining in the solution.
The maximum residue in the NaOH solution was calculated according to the SrCO3 solubility product. The NaOH solutions containing the SrCO3 precipitants were then centrifuged three times at
3000 rev min–1 for 10 min and washed in between with deionised and degassed water to remove
NaOH and to reach a pH of 7. Keeping the tubes opened for washing as briefly as possible prevented contamination by atmospheric CO2 during sample preparation. After washing, the remaining
water was removed from the vials and the SrCO3 was dried at 105°C. The SrCO3 was analysed on
the IRMS for δ13C values. For the microbial biomass, an aliquot of the K2SO4 samples was pipetted
directly into tin capsules and dried at 60°C prior to IRMS analyses.
Calculations
The 14C activity found in a certain compartment (act, i.e. shoot, root, soil, CO2, or microbial biomass) was related to the total 14C recovery after sampling, i.e. the sum of 14C activity in shoots
(ashoot), roots (aroot), soil (asoil), and CO2 (aCO2), and was termed 14Cct (data are shown in Fig. 4-3):
14
Cct =
ashoot
act
⋅ 100%
+ aroot + asoil + aCO2
(4-1).
Root-derived carbon in CO2 and microbial biomass were calculated based on 14C activity in the
plant shoots (14Cshoot); the total amount of carbon in the shoots (Cshoot) and the 14C activity in CO2
(14CCO2) and microbial biomass (14CMB) according to Kuzyakov et al. (1999)(data are shown in
Table 4-1 and Fig. 4-3):
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
C maize−ct = C shoot ⋅
67
14
14
C ct
C shoot
(4-2),
where Cmaize–ct is the amount of maize-derived carbon in a compartment (CO2 or microbial biomass)
and 14Cct is the 14C activity in that compartment related to the total recovery (Eq. (4-1)). The 14C
activity and the amount of C in the shoots were chosen as a reference because these can be measured more accurately in the shoots compared to the roots, where adhering soil particles increase the
root mass in all replicates. The amount of maize-derived carbon (Cmaize–ct) was then related to the
amount of total carbon in a compartment (Ctotal–ct) and was termed fmaize–ct:
f maize−ct =
C maize−ct
⋅100%
Ctotal −ct
(4-3).
A mass balance equation was used to determine the δ13C value of microbial biomass (δ13CMB):
δ C MB =
13
δ 13C fum ⋅ C fum −δ 13C nf ⋅ C nf
C fum − C nf
(4-4),
where δ13Cfum and δ13Cnf are the δ13C values of the fumigated and non-fumigated samples, respectively, and Cfum and Cnf are the amounts of C in the fumigated and non-fumigated K2SO4 samples,
respectively.
In the beginning of every CO2 trapping, a small volume of atmospheric CO2 was present in the
closed system, especially in the soil pore space and in the trapping tube above the NaOH solution.
We eliminated this atmospheric CO2 from the measured δ13C value using a mass balance equation:
13
δ CCO2
δ 13Ctotal ⋅ Ctotal −δ 13C air ⋅ C air
=
Ctotal − C air
(4-5),
where δ13CCO2 is the corrected δ13C value of soil air without atmospheric air, δ13Ctotal is the measured δ13C value of CO2, δ13Cair is the δ13C value of ambient air (–7.8‰, see Boutton (1991)), Ctotal
is the amount of CO2-C trapped in NaOH, and Cair is the amount of C in the soil pore space and the
trapping tube in our closed system (0.024 mg C) calculated from a CO2 concentration of 345 mg
kg–1 (Boutton, 1991) and the volume of air in the system.
After having calculated the δ13C of microbial biomass (Eq. (4-4)) and the corrected δ13C of total
CO2 efflux (Eq. (4-5)), it became possible to calculate the contributions of the C4 plant source carbon to below-ground CO2 (fC4–CO2) and to microbial biomass (fC4–MB):
f C4 −CO2 =
f C4 − MB =
δ 13 CCO2 − δ 13 C SOM
⋅100%
(4-6),
δ 13 C MB − δ 13 C SOM
⋅ 100%
δ 13 C maize − δ 13 C SOM
(4-7),
δ 13 C maize − δ 13 C SOM
68
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
where δ13Cmaize is the δ13C value of maize roots and δ13CSOM is the δ13C value of SOM from unplanted soil (data are shown in Table 4-3).
Isotopic fractionations were considered between SOM and SOM-derived CO2, between SOM
and microbial biomass, and between rhizodeposits and microbial biomass. The fractionation between maize rhizodeposits and microbial biomass was assumed to be the same as the fractionation
between SOM and microbial biomass. Since the δ13C value of root-derived CO2 is dominated by the
δ13C value of CO2 from root-respiration, we assumed no 13C fractionation between root-derived
carbon and CO2 according to Werth and Kuzyakov (2006). Considering these fractionations,
δ13CSOM in Eq. (4-6) was replaced by δ13CSOM–CO2, δ13CSOM in Eq. (4-7) was replaced by δ13CSOM–MB,
and δ13Cmaize in Eq. (4-7) was replaced by δ13Cmaize–MB:
δ 13 C SOM −CO2 = δ 13 C SOM + ε SOM −CO2
(4-8),
δ 13 C SOM − MB = δ 13 C SOM + ε SOM − MB
(4-9),
δ 13 C maize− MB = δ 13 C maize + ε SOM − MB
(4-10),
where εSOM–CO2 and εSOM–MB are 13C isotopic fractionations as absolute values in ‰ between SOM
and CO2 and between SOM and microbial biomass from unplanted soil, respectively.
Standard deviations (SD) were calculated as a variability parameter for all our results. We used a
one-way analysis of variance (ANOVA) to identify differences between δ13C values of various below-ground carbon pools, between 14C recoveries at the five sampling dates in a certain pool, between maize-derived CO2 contributions calculated by 14C or 13C tracers at a certain sampling date,
and between maize-derived carbon contributions to the microbial biomass calculated by 14C or 13C
tracers at a certain sampling date. A Fisher LSD test was used as post hoc test to identify individual
differences. Where variances were not equal, a Studentised maximum modulus test had to be applied as post hoc test. Statistics were calculated with the SPSS 10.0 package.
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
69
Results
Plant and soil carbon pools
Between days 16 and 34 the amount of carbon in the maize shoots increased linearly by 33.4 mg C
d–1 (Fig. 4-2a). Continuing this linear trend would lead to 1057 mg C on average in the shoots on
day 40, but the actual amount of carbon was about 300 mg higher on the last sampling date (Table
4-1). Hence, the shoot biomass was no longer increasing linearly between days 34 and 40 (but
rather exponentially). The maize roots grew linearly and gained 142 mg C within the whole sampling period of 24 days (Fig. 4-2b). Such a linear increase of shoot and root biomass is a prerequisite for calculating the root-derived carbon contributions to the microbial biomass and the CO2
efflux by Eq. (4-3).
The amount of carbon in the soil planted with maize was constant during the whole growth period, averaging 5784 mg C (Table 4-1). Although the roots were growing and increasing amounts of
rhizodeposits should have been supplied, the amount of carbon in the microbial biomass was also
constant at 86 mg C on average from days 22 to 40 (Table 4-1). On the first sampling date, however, the amount of carbon in the microbial biomass was significantly higher (P < 0.05) compared
to the following dates. Significant differences could not be tested between the maize soil and the
unplanted soil because only one to two soil samples per date were available for the unplanted soil.
Since total carbon of the unplanted soil was always within the standard deviation of the related
sample in the planted soil, no significant difference between the two treatments can be assumed. A
similar relationship between planted and unplanted treatments was found for the amounts of carbon
in the microbial biomass. The cumulative CO2 efflux from the planted soil increased linearly from
Fig. 4-2: Linear regressions of amounts of carbon (closed diamonds) in (a) maize shoots and (b) maize roots
towards time of maize growth (n = 4). The amounts of carbon on day 40 in the shoots (open diamonds) are
not included in the linear function.
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
70
8.5 mg C d–1 between days 22 to 28 to 12.3 mg C d–1 between days 34 and 40 (Table 4-1). In contrast, the control pots without plants showed a reduced rate of increase (from 1.4 mg C d–1 minimum between days 34 and 40 to 5.4 mg C d–1 maximum between days 24 and 34).
Table 4-1: Carbon in shoots, roots, soil, cumulative CO2 efflux, and microbial biomass on five sampling
dates of maize grown on C3 soil (means ± SD, n = 4) and of unplanted C3 soil (means ± SD, 1 ≤ n ≤ 2).
C [mg]
Days of maize
growth
maize on C3 soil
16
22
28
34
40
unplanted C3 soil
16
22
28
34
40
shoots
265
451
634
872
1329
±
±
±
±
±
12
31
46
97
66
n.a.
n.a.
n.a.
n.a.
n.a.
roots
86
108
153
198
228
±
±
±
±
±
11
11
7
23
6
n.a.
n.a.
n.a.
n.a.
n.a.
soil
5886
5831
5697
5846
5659
±
±
±
±
±
211
198
944
190
205
5972
5930
6003
5857 ± 3
5651 ± 175
CO2
111
181
232
286
359
±
±
±
±
±
microbial
biomass
5
17
11
11
9
125
80
98
84
83
63
72
80
113 ± 11
121 ± 5
122
81
68
74
83
±
±
±
±
±
13
13
9
11
4
n.a.: not applicable
CO2 was trapped for 7, 14, 21, 28, and 35 days on sampling days 16, 22, 28, 34, and 40, respectively
14
C activities
The mean 14C activities recovered from the inputs per plant from sampling dates 16 to 40 were:
62.1±17.8%, 47.8±9.7%, 37.7±4.8%, 35.7±5.8%, and 36.0±2.6% (Table 4-2). After every additional 14C pulse, the total radioactivity, however, was increasing in all pools. Most 14C was allocated
to the maize shoots. A maximum of 9 kBq 14C was translocated into the soil at the end of the experiment. The 14C activity in the soil microbial biomass made up about one third at maximum of the
14
C activity in the soil. The loss of 14C label by shoot respiration was increasing from one third to
two thirds of the input until the end of the experiment. The partitioning of 14C activity into the five
different pools in relation to the total recovery was constant throughout the experiment (Fig. 4-3). It
amounted on average to 67.1±1.6% for shoots, 10.2±0.8% for roots, 1.9±0.6% for the soil,
20.7±1.4% for the CO2 efflux, and 0.5±0.3% for the microbial biomass (the latter not shown in Fig.
4-3). Only on the first sampling there was significantly more 14C in the soil than at the other sampling dates (P < 0.05).
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
71
Table 4-2: Average 14C activity (n = 4) per plant container in plant and soil pools after repeated labelling of
maize shoots in a 14CO2 atmosphere.
Days of maize growth
Days of
14
14
C pulse labelling
16
22
28
34
40
9
9, 15
9, 15, 21
9, 15, 21, 27
9, 15, 21, 27, 33
C activity [kBq]
Total input
246.4
493.1
739.7
986.4
Shoots
Roots
Soil
CO2
Sum of recovery
102.3 ± 44.8
16.0 ± 3.5
4.1 ± 0.6
163.9 ± 46.5
23.8 ± 2.2
4.1 ± 0.3
182.7 ± 25.1
29.2 ± 2.8
3.4 ± 0.6
238.7 ± 39.9
36.9 ± 6.1
6.6 ± 2.2
302.4 ± 31.0
39.5 ± 3.4
8.8 ± 1.2
30.6 ± 19.5
153.1 ± 43.9
43.9 ± 7.7
235.5 ± 47.9
63.2 ± 8.5
278.6 ± 35.5
70.0 ± 19.0
352.1 ± 57.4
93.5 ± 1.8
444.1 ± 32.4
1.3 ± 0.3
93.4 ± 43.9
0.5 ± 0.2
257.5 ± 47.9
0.9 ± 0.2
461.2 ± 35.5
2.1 ± 0.6
634.2 ± 57.4
2.0 ± 0.6
788.9 ± 32.4
Microbial biomass
Loss by shoot respiration
1233.0
Fig. 4-3: Partitioning of 14C activity into maize shoots (hatched shading), roots (white shading), soil (black
shading), and CO2 efflux (dotted shading). Shoots were consecutively pulse-labelled, initially on day 16 and
finally on day 40 (total 5 pulses). Values are means (n = 4) with standard deviations shown to one side of the
bars only. Significant differences within one type of pool are labelled as *, i.e. P < 0.05.
δ13C values
Between days 16 and 40, the δ13C values of maize shoots and roots decreased significantly (P <
0.001), by 1.0‰ for the shoots and by 1.3‰ for the roots (Table 4-3). The δ13C of the total CO2
efflux from planted soil (–16.9‰ on average over time) was, by 1‰, significantly more negative (P
< 0.05) than the δ13C of the roots. Nevertheless, δ13C values of roots and CO2 were very close. The
δ13C values of CO2 presented in Table 4-3 were corrected by Eq. (4-5) for small amounts of air CO2
72
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
remaining in the soil pores and in the trapping tube. This correction made the δ13C values of belowground CO2 slightly more negative compared to uncorrected data, but this difference was less than
0.02‰. The δ13C of SOM was constant and amounted to –26.8‰. Until day 40, the δ13C of microbial biomass increased from –24.6‰ to –22.5‰; the mean value was –23.7‰, which was significantly more positive than the mean δ13C value of SOM (P < 0.001).
The average δ13C of SOM in unplanted soil (–27.0‰) was the same as in planted soil (Table
4-3). In the total CO2 efflux of unplanted soil, the mean δ13C between days 22 and 40 was –21.8‰.
The mean δ13C of microbial biomass between days 22 and 40 was intermediate between these two
values (–23.8‰). Consequently, there was a 13C fractionation of about 3.2‰ between organic matter in unplanted soil and microbial biomass (εSOM–MB, P < 0.001), and of 2.0‰ between microbial
biomass and microbially respired CO2 (P < 0.05). The fractionation εSOM–CO2 between SOM and
microbial CO2 was 5.2‰ (P < 0.001).
Table 4-3: δ13C values of shoots, roots, soil, CO2 efflux, and microbial biomass on five sampling dates of
maize grown on C3 soil (means ± SD, n = 4) and of unplanted C3 soil (means ± SD, 1 ≤ n ≤ 2).
δ13C [‰]
Days of maize
growth
maize on C3 soil
16
22
28
34
40
unplanted C3 soil
16
22
28
34
40
shoots
-15.2
-15.8
-16.2
-16.0
-16.2
±
±
±
±
±
0.1
0.1
0.1
0.1
0.1
n.a.
n.a.
n.a.
n.a.
n.a.
roots
-14.9
-15.7
-16.0
-16.1
-16.2
±
±
±
±
±
0.2
0.3
0.1
0.2
0.2
n.a.
n.a.
n.a.
n.a.
n.a.
soil
-26.7
-26.9
-26.9
-26.5
-26.7
±
±
±
±
±
CO2
0.4
0.1
0.1
0.4
0.1
-16.9
-16.7
-17.9
-16.8
-16.7
-26.8
-27.0
-27.3
-26.9 ± 0.1
-26.8 ± 0.1
-18.6
-21.9
-21.2
-21.7
-22.6
±
±
±
±
±
0.0
0.5
0.2
0.3
0.8
microbial
biomass
-24.6
-25.1
-23.4
-23.0
-22.5
±
±
±
±
±
0.9
0.2
0.3
0.5
0.7
n.d.
-23.0
-23.1
-24.9 ± 0.6
-24.2
n.a.: not applicable
n.d.: not determined
Contributions of maize roots to CO2 efflux and microbial biomass
The comparisons between the two different methods for calculating the contributions of rootderived CO2 to total CO2 efflux showed the following results (Fig. 4-4): First, no significant difference was found on day 16 between these contributions as calculated by the 14C approach and the
13
C approach with and without fractionation. Second, only the 13C approach without fractionation
yielded significantly more root-derived CO2 (91% on day 22, 94% on day 34) versus the 14C approach (P < 0.05). Consideration of 13C fractionation between SOM and CO2 led to equal percentages of root-derived CO2 on those days. Third, the results on day 28 from both 13C methods were
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
73
significantly smaller (without 13C fractionation P < 0.05, with 13C fractionation P < 0.001) than the
94% calculated by the 14C approach. Fourth, the root-derived CO2 contribution based on the 14C
method exceeded the 100% level by 16% on the last sampling day. The result from the 13C method
without fractionation was below 100%, but not significantly different from the 14C result. Considering the 13C fractionation led to a significantly smaller root-derived CO2 contribution (91%) than the
14
C approach (P < 0.05).
Fig. 4-4: Contributions of root-derived CO2 to total CO2 efflux from a C3 soil planted with maize. Methods
used to calculate the root-derived CO2 contributions are: the 14C labelling technique (hatched shading), the
natural 13C labelling technique (dotted shading), and the natural 13C labelling technique with a fractionation
of 5.2‰ between SOM and CO2 (black shading). Error bars show standard deviations (n = 4). On each day,
significant differences are shown for the natural 13C labelling technique with and without fractionation when
compared with the 14C labelling technique (*P < 0.05, ***P < 0.001).
The contributions of root-derived C to microbial biomass C calculated by the two methods increased with the age of the maize (Fig. 4-5). Using 14C labelling, this increase was significant on
day 34 (P < 0.05). On the first two sampling dates, the 13C approach without fractionation yielded
significantly higher values – up to eight times as high – than the 14C approach. Incorporating 13C
fractionation on those two days led to negative values, which were significantly different to the 14C
approach. On the last three sampling dates, the contribution calculated by the 13C approach without
fractionation was from three to eleven times as high as the contribution calculated by the 14C approach. There was no significant difference to the 14C method when 13C fractionation was considered on the last three sampling dates.
74
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
Fig. 4-5: Contributions of root-derived carbon to total microbial biomass C from a C3 soil planted with
maize. Methods used to calculate the root-derived carbon contributions are: the 14C labelling technique
(hatched shading), the natural 13C labelling technique (dotted shading), and the natural 13C labelling technique with a fractionation of 3.2‰ between SOM and microbial biomass and between rhizodeposits and
microbial biomass (black shading). Error bars show standard deviations (n = 4). On each day, significant
differences are shown for the natural 13C labelling technique with and without fractionation when compared
with the 14C labelling technique (*P < 0.05, **P < 0.01, ***P < 0.001).
Discussion
Comparison of the 14C pulse labelling and natural 13C labelling approaches to estimate
maize-root-derived carbon contributions
Both methods showed similar contributions to total CO2 efflux from the maize rhizosphere when
13
C fractionations between SOM and CO2 were considered (Fig. 4-4). In a previous publication on
maize grown on C3 soil (Werth et al., 2006), we concluded that the natural 13C labelling technique
overestimated root respiration by comparing observed CO2 partitioning into three sources with literature results. By validating the 13C results with 14C results, we now determined that the rootderived carbon was between 69 and 94% of below-ground CO2 efflux. Thus, the results from Werth
et al. (2006) were correct rather than an overestimation. Under controlled conditions, other comparable studies report root-derived carbon values about 75% for 38-day-old spring wheat (Triticum
aestivum L.) (Kuzyakov and Cheng, 2001) or between 35 and 41% for 42-day-old ryegrass (Lolium
perenne L.) (Chen et al., 2006). Our values therefore correspond with the upper range of these studies. The broad range of root-derived CO2 already mentioned in the introduction reveals, however,
that CO2 efflux partitioning very much depends on the observed plants, their growth stage, con-
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
75
trolled or field conditions, soil preparation etc. This problem always has to be considered when
comparisons between different studies are made.
The exceeding of 100% of the total CO2 efflux on day 40 using the 14C labelling technique reflects non-linear plant growth up to that day (Table 4-1, Fig. 4-2). The linear model of Kuzyakov et
al. (1999) to calculate root-derived C amounts from 14C activity was not applicable on day 40. Had
the plants grown linearly up to day 40, the root-derived contribution would have been 91% – another comparable result to 13C labelling including fractionation. Hence, 116% on day 40 (Fig. 4-4)
is an error and should not be compared to the 13C method.
The airflow in our tubing system enabled us not only to trap CO2, it also assured against leakage
of CO2 out of the system or into the PVC tube’s wall. We chose PVC tubes since they generally are
airtight. Small errors in our CO2 budget could have arisen, however, by a minimum of diffusion
through the walls (either into or out of the tube). If these errors were present, they would affect both
methods in the total amount of CO2, i.e. 14C pulse labelling in Eq. (4-3) and natural 13C labelling in
Eq. (4-5).
On the last three sampling dates, the 14C and 13C techniques showed similar results for the rootderived carbon in the microbial biomass carbon (when 13C fractionation between SOM and microbial biomass was considered; Fig. 4-5). Without fractionation, values were significantly different to
the 14C approach on all sampling dates. Hence, like for the CO2 efflux, fractionations should be
considered. Root-derived C contributions to total microbial biomass C ranging from 1 to 11% after
42 days of maize growth (Qian and Doran, 1996), from 8 to 10% within one growth period
(Rochette et al., 1999), or up to 23% after a single year of maize growth (Bruulsema and Duxbury,
1996) confirm our findings of 2 to 11% with 13C fractionation (Fig. 4-5). Our previous publication
showed that – without fractionation between the substrate and the CO2 in calculating the δ13C value
of microbial CO2 – about 37% of the microbial biomass in the rhizosphere was active, i.e. feeding
on a C4 source (Werth et al., 2006). In that study, also assuming a 5‰ fractionation in the microbial
substrate respiration when calculating the δ13C of microbial respiration would reduce the amount of
C4-derived C in the microbial biomass to 9%. This closely corresponds to the 7 to 11% root-derived
C in the microbial biomass on the last two sampling dates of the present study.
Soil samples for microbial biomass extraction were prepared by hand-picking roots. Small
amounts of fine roots could still have been present in the samples resulting in destruction of their
cell membranes after chloroform fumigation and a contribution of the cell content to the microbial
biomass extract. Consequently, total carbon in microbial biomass, 14C activity, and C4 plant contribution to the δ13C value could have been overestimated. This source of error could be overcome by
a more complex preparation of soil microbial biomass samples, e.g. by pre-extraction with K2SO4,
wet sieving or centrifuging (Mueller et al., 1992). The same error, however, has to be considered in
all the other studies compared to the present experiment.
To be consistent in the assumption of linear shoot growth (used when calculating CO2 efflux partitioning), we also consider such growth up to day 40 when calculating root-derived carbon in the
76
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
microbial biomass by the 14C technique: this assumption yields a decrease from 14 to 9%. This result would be closer to the root-derived C contribution calculated by the 13C approach with fractionation. Hence, due to non-linear plant growth, the 14C approach should not be used between day
34 and 40. On days 16 and 22, 13C with fractionation yielded negative values. This reflects the high
average fractionation (3.2‰) between SOM and microbial biomass, leading to more positive δ13C
values of SOM than of microbial biomass. Using the maximum possible fractionation εSOM–MB in the
maize treatment (2.2‰ on day 16 and 1.9‰ on day 22), the contribution of root-derived carbon to
the microbial biomass on days 16 and 22 would be zero. This result would again closely reflect the
14
C result. It is therefore important to determine the actual fractionation on every sampling day –
and not an average fractionation over time – with an appropriate number of replicates and consequently include time changes of the fractionation.
While both methods – 14C pulse labelling and natural 13C labelling – worked sufficiently in the
determination of root-derived C in the CO2 efflux from soil, this was much more problematic with
the microbial biomass. Although chloroform fumigation-extraction has become a standard method
in soil biology, the absence of plant effect on microbial biomass size (Table 4-1) and δ13C (Table
4-3) makes an estimation of root-derived C to microbial biomass very uncertain using this coarse
method. As a substitute, molecular methods like biomarkers could be used in combination with isotopic tracer methods to identify C4-derived C contributions to certain community parts of the soil
microbial biomass (δ13C of phospholipid fatty acids (PLFA)), or to determine plant-derived or microbial residues (individual sugars) (reviewed by Glaser, 2005).
Advantages and limitations of the 14C pulse labelling approach
General advantages and limitations of the 14C method are presented in Table 4-4. The method of
transforming 14C activity into amounts of carbon in particular pools has been used before (Kuzyakov et al., 1999; Kuzyakov and Cheng, 2001; Kuzyakov et al., 2001; Kuzyakov et al., 2003). A
similar method that also considered small increases of plant biomass between individual labelling
pulses was suggested by Remus et al. (2006). Our calculation method, however, allows only a
rough estimate of the C passed through each flow because the parameters of Eq. (4-2) are not constant during plant development. This method can be used only after completed 14C distribution in
the plant. For grasses and cereals, this completion takes five days after assimilation (Domanski et
al., 2001). In accordance to that study, the distribution of the 14C tracer between the plant-soil compartments was completed on all sampling days in our study, since there were no significant differences between sampling days (Fig. 4-3). Hence, the equation was applicable in our study. Due to
non-linear carbon assimilation after day 34, however, the linear equation can no longer be applied at
the end of the growth period.
Another limitation of the 14C method is that pulse labelling can effectively track inputs derived
from recent assimilates, but that this input is unlikely to constitute the most abundant source of substrate to microbial communities in the rhizosphere (Thornton et al., 2004). Non-recent assimilates,
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
77
Table 4-4: Advantages and limitations of 14C pulse labelling and natural 13C labelling techniques for estimating the contribution of root-derived C to the total CO2 efflux from soil and to the soil microbial biomass.
14
C pulse labelling
-
Advantages
-
Limitations
-
high sensitivity of the contribution of
plant-derived C to CO2 and to microbial
biomass
information on distribution of assimilated
C in individual stages of plant development
allows estimating the incorporation of
plant C into pools with low and very low
turnover rates
one or many pulses are possible
easy to handle
cheap purchase costs and individual
analyses
uncompleted distribution of labelled C
between plant organs and below-ground
pools if sampling is done too early after
the labelling
recalculation of total rhizodeposition is
suitable only for linear growth periods
provides only distribution of recently
assimilated C at specific development
stages of plants
both non-recent and recent assimilates
can be traced if labelling pulses are repeated
no recalculation of distribution to whole
growth period
radioactivity hazards
laborious labelling sessions with chambers required
Natural 13C labelling
-
-
continuous labelling of plants and soil
pools
no labelling equipment required
no radioactivity precautions necessary
easy usage under laboratory and field
conditions
very low sensitivity of the contribution of
plant-derived C to CO2 and to microbial
biomass
only incorporation of plant-derived C
into pools with high turnover rates during one vegetation period is possible
applicable only on pure C3 or C4 soil
contamination with air CO2
high variation of δ13C in CO2 or microbial biomass possible
results are strongly affected by 13C fractionation
results are strongly affected by
preferential isotope utilization
expensive purchase costs and individual
analyses
i.e. more complex organic forms, will be exuded much later than recent assimilates; they will also
be processed at different rates and most likely by different microorganisms. This problem, however,
can be overcome by a series of 14C labelling pulses as used in our study or by continuous labelling
techniques, like natural 13C labelling. The triplication of root-derived carbon in the microbial biomass on day 34 (Fig. 4-5) indicates that both recent and non-recent assimilates contribute to the
root-derived carbon in the microbial biomass from that day onwards.
Advantages and limitations of the natural 13C labelling approach
In contrast to the artificial 14C labelling technique, the natural 13C labelling approach corresponds to
continuous labelling of plants and plant-derived soil pools (Table 4-4). A major limitation is that
78
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
four assumptions are involved concerning 13C isotopic effects during root- and SOM-derived respiration and during utilization of rhizodeposits and SOM by the microbial biomass:
1) The δ13C isotope signature of root-derived CO2 is the same as the δ13C value of the
roots.
2) The δ13C isotope signature of SOM-derived CO2 equals the δ13C value of SOM.
3) The δ13C isotope signature of root-derived microbial biomass corresponds to the δ13C
value of roots and rhizodeposits.
4) The δ13C isotope signature of SOM-derived microbial biomass is equal to the δ13C value
of SOM.
According to Werth and Kuzyakov (2006) we can only accept the first assumption, since the
13
δ C value of root-derived CO2 is dominated by the δ13C value of CO2 from root respiration. Our
unplanted control treatment refutes the last three assumptions. We had to consider mean 13C fractionations of 3.2‰ between SOM and microbial biomass and of 5.2‰ between SOM and microbially respired CO2. Henn and Chapela (2000) have shown that the 13C fractionation differs during
decomposition of C3- and C4-derived sucrose by three specific fungi. However, we assumed the
fractionation between maize rhizodeposits and microbial biomass in Eqs. (4-7) and (4-10) to be
equal to the fractionation between SOM and microbial biomass (εSOM–MB = 3.2‰). In line with earlier studies (Balesdent and Mariotti, 1996; Boutton, 1996; Bol et al., 2003), we accepted this assumption because we had no direct measure to determine the actual fractionation between
rhizodeposits and the microbial biomass. This determination is a future challenge, requiring that
rhizodeposits be decomposed by exactly the same microbial community as developed in our C4plant-containers. In the present study, however, we assumed equal fractionations for C3- and C4derived substrates because the root-derived contributions calculated with and without 13C fractionation for the C4 substrate were not significantly different. In Fig. 4-5, the latter would read –11, –17,
8, 9, and 16% on sampling dates 16 to 40, respectively.
The fractionations between SOM and microbial biomass and between SOM and CO2 do not only
include isotopic effects per se. They also include preferential utilization of substrates with different
biological availability and different δ13C values. The first fractionation step leading to a 13Cenriched microbial biomass compared to SOM can be explained by isotope discrimination during
biosynthesis of new microbial biomass (Potthoff et al., 2003). Compared to SOM, water-soluble
organic compounds with a heavier isotopic composition are preferentially used by soil microorganisms (Henn and Chapela, 2000; Pelz et al., 2005). The second fractionation step yields more 13Cenriched microbial CO2 compared to the microbial biomass and the substrate. Usually, CO2 from
microbial respiration is 13C-depleted compared to the feeding substrate (Blair et al., 1985; Mary et
al., 1992; Potthoff et al., 2003). Opposite results, i.e. 13C enrichment of CO2 versus source, can be
explained by a selective use of 13C-enriched SOM compounds by microorganisms (Ågren et al.,
1996; Werth et al., 2006). This selection was more pronounced than the 13C depletion effect of the
metabolism itself (Šantrůčková et al., 2000), resulting in 13C-enriched CO2. The δ13C value of CO2
changes during increasing decomposition of plant residues by –5 to +2‰ compared to the δ13C
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
79
value of the original substrate (Hamer et al., 2004). This requires considering both average fractionations on a single sampling date and their changes during a study. Total fractionation between
SOM and microbial biomass and between SOM and CO2 – including kinetic fractionation and preferential utilization – is important when using 13C natural abundance methods. Compared to other
pools like plants or bulk soil, the δ13C values of CO2 and microbial biomass often have the highest
variation, which will affect the accuracy of below-ground carbon partitioning. Hence, fractionations
should be determined with a large number of replicates to ensure exact calculations of plant-derived
C to the CO2 efflux and to the soil microbial biomass. Otherwise, problems could occur like the
negative root-derived C contributions to the microbial biomass reported in this study.
Conclusions
The 14C pulse labelling technique and the natural 13C labelling technique yielded similar contributions of root-derived carbon to the CO2 efflux from soil, when 13C fractionation in the latter approach was considered between SOM and CO2. Both methods also yielded similar contributions of
root-derived carbon to the microbial biomass when 13C fractionation between SOM and microbial
biomass was considered. This calls for determining 13C fractionation in calculations of maizederived carbon contributions. Rhizodeposition and root-derived CO2 efflux should only be estimated by the 14C labelling method when plant biomass increases linearly.
Acknowledgements
The German Research Foundation (DFG) supported this study. The authors would also like to thank
I. Subbotina and G. Bermejo-Dominguez for their practical support during the experiment, Dr. V.
Cercasov for usage of the scintillation counter, and Dr. W. Armbruster and E. Dachtler for the
IRMS analyses.
References
Ågren, G.I., Bosatta, E., Balesdent, J., 1996. Isotope discrimination during decomposition of organic matter: a theoretical analysis. Soil Science Society of America Journal 60, 1121-1126.
Andrews, J.A., Harrison, K.G., Matamala, R., Schlesinger, W.H., 1999. Separation of root respiration from total soil respiration using carbon-13 labeling during free-air carbon dioxide
enrichment (FACE). Soil Science Society of America Journal 63, 1429-1435.
Angers, D.A., Voroney, R.P., Côté, D., 1995. Dynamics of soil organic matter and corn residues
affected by tillage practices. Soil Science Society of America Journal 59, 1311-1315.
Balesdent, J., Mariotti, A., 1996. Measurement of soil organic matter turnover using 13C natural
abundance. In: Boutton, T.W., Yamasaki, S.I. (Eds.), Mass Spectrometry of Soils, Marcel
Dekker, New York, USA, pp. 83-111.
80
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
Blair, N., Leu, A., Munoz, E., Olsen, J., Kwong, E., Des Marais, D., 1985. Carbon isotope fractionation in heterotrophic microbial metabolism. Applied and Environmental Microbiology
50, 996-1001.
Bol, R., Moering, J., Kuzyakov, Y., Amelung, W., 2003. Quantification of priming and CO2 respiration sources following slurry-C incorporation into two grassland soils with different C
content. Rapid Communications in Mass Spectrometry 17, 1-6.
Boutton, T.W., 1991. Stable carbon isotope ratios of natural materials: II. atmospheric, terrestrial,
marine, and freshwater environments. In: Coleman, D.C., Fry, B. (Eds.), Carbon Isotope
Techniques. Isotopic Techniques in Plant, Soil, and Aquatic Biology, Academic Press, Inc.,
San Diego, pp. 173-185.
Boutton, T.W., 1996. Stable carbon isotope ratios of soil organic matter and their use as indicators
of vegetation and climate change. In: Boutton, T.W., Yamasaki, S.I. (Eds.), Mass
Spectrometry of Soils, Marcel Dekker, New York, pp. 47-82.
Bruulsema, T.W., Duxbury, J.M., 1996. Simultaneous measurement of soil microbial nitrogen,
carbon, and carbon isotope ratio. Soil Science Society of America Journal 60, 1787-1791.
Chen, C.R., Condron, L.M., Xu, Z.H., Davis, M.R., Sherlock, R.R., 2006. Root, rhizosphere and
root-free respiration in soils under grassland and forest plants. European Journal of Soil Science 57, 58-66.
Cheng, W., 1996. Measurement of rhizosphere respiration and organic matter decomposition using
natural 13C. Plant and Soil 183, 263-268.
Cheng, W., Coleman, D.C., Carroll, C.R., Hoffman, C.A., 1993. In situ measurement of root
respiration and soluble C concentrations in the rhizosphere. Soil Biology & Biochemistry 25,
1189-1196.
Domanski, G., Kuzyakov, Y., Siniakina, S.V., Stahr, K., 2001. Carbon flows in the rhizosphere
of ryegrass (Lolium perenne). Journal of Plant Nutrition and Soil Science 164, 381-387.
Ekblad, A., Högberg, P., 2000. Analysis of δ13C of CO2 distinguishes between microbial respiration of added C4-sucrose and other soil respiration in a C3-ecosystem. Plant and Soil 219,
197-209.
Ekblad, A., Högberg, P., 2001. Natural abundance of 13C in CO2 respired from forest soils reveals
speed of link between tree photosynthesis and root respiration. Oecologia 127, 305-308.
Ekblad, A., Nyberg, G., Högberg, P., 2002. 13C-discrimination during microbial respiration of
added C3-, C4- and 13C-labelled sugars to a C3-forest soil. Oecologia 131, 245-249.
Fernandez, I., Cadisch, G., 2003. Discrimination against C-13 during degradation of simple and
complex substrates by two white rot fungi. Rapid Communications in Mass Spectrometry 17,
2614-2620.
Glaser, B., 2005. Compound-specific stable-isotope 13C analysis in soil science. Journal of Plant
Nutrition and Soil Science 168, 633-648.
Grace, J., 2004. Understanding and managing the global carbon cycle. Journal of Ecology 92, 189202.
Gregorich, E.G., Liang, B.C., Drury, C.F., Mackenzie, A.F., McGill, W.B., 2000. Elucidation of
the source and turnover of water soluble and microbial biomass carbon in agricultural soils.
Soil Biology & Biochemistry 32, 581-587.
Hamer, U., Dalhus, W., Marschner, B., Schulte, U., Gleixner, G., 2004. Isotopic 13C fractionation during the mineralisation of organic substrates. In: Hamer, U. (Ed.) Priming effects of
dissolved organic substrates on the mineralisation of lignin, peat, soil organic matter and
black carbon determined with 14C and 13C isotope techniques, PhD thesis at the RuhrUniversität Bochum, Germany, pp. 129-150.
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
81
Henn, M.R., Chapela, I.H., 2000. Differential C isotope discrimination by fungi during decomposition of C3- and C4-derived sucrose. Applied and Environmental Microbiology 66, 41804186.
Kelting, D.L., Burger, J.A., Edwards, G.S., 1998. Estimating root respiration, microbial respiration in the rhizosphere, and root-free soil respiration in forest soils. Soil Biology & Biochemistry 30, 961-968.
Kristiansen, S.M., Brandt, M., Hansen, E.M., Magid, J., Christensen, B.T., 2004. 13C signature
of CO2 evolved from incubated maize residues and maize-derived sheep faeces. Soil Biology
& Biochemistry 36, 99-105.
Kuzyakov, Y., Cheng, W., 2001. Photosynthesis controls of rhizosphere respiration and organic
matter decomposition. Soil Biology & Biochemistry 33, 1915-1925.
Kuzyakov, Y., Cheng, W., 2004. Photosynthesis controls of CO2 efflux from maize rhizosphere.
Plant and Soil 263, 85-99.
Kuzyakov, Y., Kretzschmar, A., Stahr, K., 1999. Contribution of Lolium perenne rhizodeposition
to carbon turnover of pasture soil. Plant and Soil 213, 127-136.
Kuzyakov, Y., Ehrensberger, H., Stahr, K., 2001. Carbon partitioning and below-ground translocation by Lolium perenne. Soil Biology & Biochemistry 33, 61-74.
Kuzyakov, Y., Raskatov, A., Kaupenjohann, M., 2003. Turnover and distribution of root exudates of Zea mays. Plant and Soil 254, 317-327.
Liang, B.C., Wang, X.L., Ma, B.L., 2002. Maize root-induced change in soil organic carbon
pools. Soil Science Society of America Journal 66, 845-847.
Lin, G., Rygiewicz, P.T., Ehleringer, J.R., Johnson, M.G., Tingey, D.T., 2001. Time-dependent
responses of soil CO2 efflux components to elevated atmospheric [CO2] and temperature in
experimental forest mesocosms. Plant and Soil 229, 259-270.
Mary, B., Mariotti, A., Morel, J.L., 1992. Use of 13C variations at natural abundance for studying
the biodegradation of root mucilage, roots and glucose in soil. Soil Biology & Biochemistry
24, 1065-1072.
Merckx, R., Dijkstra, A., den Hartog, A., van Veen, J.A., 1987. Production of root-derived material and associated microbial growth in soil at different nutrient levels. Biology and Fertility
of Soils 5, 126-132.
Mueller, T., Joergensen, R.G., Meyer, B., 1992. Estimation of soil microbial biomass C in the
presence of living roots by fumigation-extraction. Soil Biology & Biochemistry 24, 179-181.
Pelz, O., Abraham, W.-R., Saurer, M., Siegwolf, R., Zeyer, J., 2005. Microbial assimilation of
plant-derived carbon in soil traced by isotope analysis. Biology and Fertility of Soils V41,
153-162.
Potthoff, M., Loftfield, N., Buegger, F., Wick, B., John, B., Jörgensen, R.G., Flessa, H., 2003.
The determination of δ13C in soil microbial biomass using fumigation-extraction. Soil Biology & Biochemistry 35, 947-954.
Qian, J.H., Doran, J.W., 1996. Available carbon released from crop roots during growth as determined by carbon-13 natural abundance. Soil Science Society of America Journal 60, 828831.
Qian, J.H., Doran, J.W., Walters, D.T., 1997. Maize plant contributions to root zone available
carbon and microbial transformations of nitrogen. Soil Biology & Biochemistry 29, 14511462.
Remus, R., Augustin, J., Hüve, K., Plugge, J., 2006. Dynamik des Eintrages und der Umsetzung
von Assimilat-Kohlenstoff im Boden bei Roggen und Mais: Modellierung der C-Flüsse nach
4 Root-derived carbon in soil respiration and microbial biomass determined by 14C and 13C
82
14
C-Pulsbegasung. In: Merbach, W., Gans, W., Augustin, J. (Eds.), Reaktionen und Stoffflüsse im wurzelnahen Raum – 16. Borkheider Seminar zur Ökophysiologie des Wurzelraumes.
Beiträge aus der Hallenser Pflanzenernährungsforschung, No. 11, Verlag Grauer, Beuren,
Stuttgart, pp. 40-48.
Robinson, D., Scrimgeour, C.M., 1995. The contribution of plant C to soil CO2 measured using
δ13C. Soil Biology & Biochemistry 27, 1653-1656.
Rochette, P., Flanagan, L.B., 1997. Quantifying rhizosphere respiration in a corn crop under field
conditions. Soil Science Society of America Journal 61, 466-474.
Rochette, P., Angers, D.A., Flanagan, L.B., 1999. Maize residue decomposition measurement
using soil surface carbon dioxide fluxes and natural abundance of carbon-13. Soil Science
Society of America Journal 63, 1385-1396.
Ryan, M.C., Aravena, R., 1994. Combining 13C natural abundance and fumigation-extraction
methods to investigate soil microbial biomass turnover. Soil Biology & Biochemistry 26,
1583-1585.
Šantrůčková, H., Bird, M.I., Frouz, J., Šustr, V., Tajovský, K., 2000. Natural abundance of 13C
in leaf litter as related to feeding activity of soil invertebrates and microbial mineralisation.
Soil Biology & Biochemistry 32, 1793-1797.
Schweizer, M., Fear, J., Cadisch, G., 1999. Isotopic (13C) fractionation during plant residue decomposition and its implications for soil organic matter studies. Rapid Communications in
Mass Spectrometry 13, 1284-1290.
Thornton, B., Paterson, E., Midwood, A.J., Sim, A., Pratt, S.M., 2004. Contribution of current
carbon assimilation in supplying root exudates of Lolium perenne measured using steadystate 13C labelling. Physiologia Plantarum 120, 434-441.
Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring soil microbial biomass C. Soil Biology & Biochemistry 19, 703-707.
Warembourg, F.R., Paul, E.A., 1977. Seasonal transfers of assimilated 14C in grassland: plant
production and turnover, soil and plant respiration. Soil Biology & Biochemistry 9, 295-301.
Werth, M., Kuzyakov, Y., 2005. Below-ground partitioning (14C) and isotopic fractionation (δ13C)
of carbon recently assimilated by maize. Isotopes in Environmental and Health Studies 41,
237-248.
Werth, M., Kuzyakov, Y., 2006. Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize. Plant and Soil 284, 311-325.
Werth, M., Subbotina, I., Kuzyakov, Y., 2006. Three-source partitioning of CO2 efflux from soil
planted with maize by 13C natural abundance fails due to inactive microbial biomass. Soil
Biology & Biochemistry 38, 2772-2781.
Wu, J., Jörgensen, R.G., Pommerening, B., Chaussod, R., Brookes, P.C., 1990. Measurement of
soil microbial biomass-C by fumigation-extraction – an automated procedure. Soil Biology
& Biochemistry 22, 1167-1169.
Zibilske, L.M., 1994. Carbon Mineralization. In: Weaver, R.W., Angle, S., Bottomley, P., Bezdicek, D., Smith, S., Tabatabai, A., Wollum, A. (Eds.), Methods of Soil Analysis, Part 2,
Microbiological and Biochemical Properties. Soil Science Society of America Book Series,
No. 5, Soil Sci. Soc. Am., Inc., Madison, pp. 835-864.
5
Partitioning of CO2 efflux from soil planted with maize by
natural abundance and root exclusion
Martin Werth and Yakov Kuzyakov
Journal of Plant Nutrition and Soil Science (2007), under review
includes two tables and three figures
13
C
84
5 Partitioning of CO2 efflux from soil planted with maize
Summary
Two approaches to quantitative estimates of soil CO2 efflux derived from roots (including rhizomicrobial respiration) and from microbial decomposition of soil organic matter (SOM) in the field
were compared. In the root exclusion approach, root- and SOM-derived CO2 were determined by
the total CO2 effluxes from maize (Zea mays L.) and bare fallow plots. In the natural 13C labeling
approach, maize was planted on soil with C3 vegetation history and the total CO2 efflux from soil
was subdivided by isotopic mass balance. In both approaches, CO2 from SOM decomposition
dominated 65 to 78% of the total CO2 efflux during the growth period. In the natural 13C labeling
approach, it was also possible to separate rhizomicrobial respiration from root respiration, but the
former was low compared to other studies. At the end of the growth period, when considering high
isotopic fractionations between SOM, microbial biomass, and CO2, however, root and rhizomicrobial respiration amounted to 64 and 36% of root-derived CO2, respectively. This relationship was
closer to the 50:50% partitioning described in the literature than without fractionation. Fractionation
processes of 13C must be taken into account when calculating CO2 partitioning in soil. Both methods showed the same partitioning results when 13C isotopic fractionation during microbial respiration was considered.
Abbreviations:
IRMS – isotope ratio mass spectrometer, RDR – root-derived respiration, RMR –
rhizomicrobial respiration, RR – root respiration, SD – standard deviation, SOM –
soil organic matter, SOMD – soil organic matter decomposition
5 Partitioning of CO2 efflux from soil planted with maize
85
Introduction
Partitioning the total carbon dioxide (CO2) efflux from soil is very important in order to identify
individual carbon sinks or sources. This CO2 efflux can be separated into five components (Kuzyakov, 2006): (1) root respiration, i.e. respiration of assimilates by roots of autotrophic plants, (2) rhizomicrobial respiration, i.e. respiration of rhizodeposits (exudates, lysates, mucilages etc.) by
heterotrophic microorganisms in the rhizosphere, (3) decomposition of dead plant residues by heterotrophic microorganisms, (4) priming effect, i.e. plant-induced additional (or limited) decomposition of soil organic matter by heterotrophic microorganisms, and (5) decomposition of the soil
organic matter (SOM) by heterotrophic microorganisms. In the absence of plant residues and assuming a low contribution of priming effects in fertilized agricultural soils (Cheng and Coleman,
1990; Paterson and Sim, 1999; 2000), the three main components of CO2 efflux will be (1) root
respiration, (2) rhizomicrobial respiration and (3) SOM decomposition. The sum of CO2 from root
respiration and rhizomicrobial respiration is termed ‘root-derived CO2’ and the related process
‘root-derived respiration’ (RDR).
Carbon dioxide derived from soil organic matter decomposition (SOMD) and that derived from
the roots can be quantified by isotopic labeling of plants with 13C or 14C isotopes and tracing the
label in root-derived CO2 (e.g. Warembourg and Paul, 1977; Andrews et al., 1999). The difference
between this labeled fraction and the total CO2 efflux represents SOMD. Non-isotopic methods to
separate root- from SOM-derived CO2, such as a combination of trenching and excised-root methods, have also been used (Kelting et al., 1998; Chen et al., 2006). The results vary strongly depending on plants, soils, and environmental and experimental conditions. By in situ 14C labeling of
Canadian prairie grass, Warembourg and Paul (1977) found low contributions (19%) of rootderived CO2 to the total CO2 efflux from soil. On the other hand, under controlled conditions, Chen
et al. (2006) reported very high contributions of root-derived CO2 to the total CO2 efflux, with values of up to 99% in a ryegrass (Lolium perenne L.) rhizosphere. Various studies with grass species
have found results within this range with an average root-derived CO2 contribution of 70 ± 27% in
the laboratory and 36 ± 15% in the field (Table 5-1).
It is very difficult to further differentiate between the CO2 directly derived from root respiration
and that derived from mineralization of rhizodeposits (Killham and Yeomans, 2001). This separation of root respiration (RR) and rhizomicrobial respiration (RMR) is a major challenge in quantifying rhizosphere carbon flows. Separation is important to quantify carbon sources for SOM and for
rhizosphere microorganisms, to identify respiration of autotrophic and heterotrophic organisms, and
to calculate carbon turnover by rhizosphere microorganisms (Kuzyakov, 2004). Separation results
of recent studies on various grass species reveal that root-derived CO2 consists of about 50 ± 10%
for both RR and RMR (Table 5-1).
86
5 Partitioning of CO2 efflux from soil planted with maize
Table 5-1: Contributions of root respiration (RR) and rhizomicrobial respiration (RMR) to root-derived CO2
and of root-derived respiration (RDR) and SOM decomposition (SOMD) to total CO2 efflux from various
studies with grass species.
5 Partitioning of CO2 efflux from soil planted with maize
87
Some attempts to separate root and rhizomicrobial respiration were tested with various success
(reviewed by Kuzyakov and Larionova, 2005). Most of those studies, however, were conducted
under controlled conditions. The methods were based on pulse labeling of plants in a 14CO2 atmosphere and tracing the 14CO2 dynamics (Kuzyakov et al., 1999; Kuzyakov et al., 2001; Kuzyakov
and Domanski, 2002), isotopic dilution (Cheng et al., 1993) or various treatments with 14C labeled
plants and rhizodeposits (Johansson, 1992; Swinnen, 1994). The only field studies that attempted to
separate root and rhizomicrobial respiration were based on trenching and excised roots (Kelting et
al., 1998), on shading and clipping and excised roots (Craine et al., 1999), or on root exclusion and
component integration (Larionova et al., 2006) methods. All these approaches very strongly disturb
the soil and/or the roots, making the relevance of these results questionable.
The objective of this study was to compare two approaches to quantitatively estimating (1) rootderived CO2 and (2) SOM-derived CO2 contributions to total soil respiration in the field. The two
approaches were (1) root exclusion, and (2) natural 13C labeling. In the latter approach a further
partitioning of root-derived CO2 into root and rhizomicrobial respiration was also considered.
The root exclusion technique can be subdivided into three categories (Hanson et al., 2000): (1)
root removal – roots are removed, soil is placed back in reverse order of removal, and further root
growth is prevented by barriers, (2) trenching – existing roots are severed by trenching at a plot
boundary but not removed, and a barrier is installed to inhibit future root growth, and (3) gap analysis – aboveground vegetation is removed from relatively large areas and the unplanted plot is compared to a planted plot. The term ‘gap analysis’ derives from the original method where soil
respiration rates in a forested stand were compared to clear-cut ‘gaps’ within the stand (Brumme,
1995). The method used in our study can be grouped into the category of ‘gap analysis’, but since
we had no real gaps, we prefer to use the overall term ‘root exclusion’. In the root exclusion approach used in our study, the total CO2 effluxes from maize (Zea mays) and bare fallow plots were
considered to estimate root- and SOM-derived CO2. The same type of root exclusion technique was
used earlier by Rochette et al. (1999b).
The natural 13C labeling approach, allowing separation of three CO2 sources from soil, was recently suggested by Kuzyakov (2004; 2005) and was practically tested under controlled conditions
by Werth et al. (2006). In the present study we tested the approach under field conditions. The
method is based on the natural 13C labeling technique (Balesdent and Mariotti, 1996), i.e. 13C natural abundance is used by growing C4 plants on a soil developed under C3 vegetation (‘C3 soil’) or
vice versa. Hence, the δ13C values of SOM, maize roots, microbial biomass, and total CO2 efflux
from the soil are used to determine the three fractions of CO2. These contributions of RR, RMR,
and SOMD to total soil CO2 efflux can be calculated according to the isotopic mass balance of microbial biomass and CO2 (Kuzyakov, 2004; 2005). Contributions of root-derived CO2 (RDR: the
sum of RR and RMR) and of SOM-derived CO2 (SOMD) were compared to the results of the root
exclusion approach.
88
5 Partitioning of CO2 efflux from soil planted with maize
Materials and methods
Experimental set-up
In May 2004, a maize (Zea mays L., cv. Tassilo) plot (10 m2) was established on a loamy Haplic
Luvisol from loess with C3 vegetation history (grasses with dominance of Lolium perenne L.), located on the University of Hohenheim’s research farm ‘Heidfeldhof’ in Stuttgart, Germany (48° 42´
50´´ N, 9° 11´ 21´´ E). No C4 plants have ever been planted on this plot before, which was crucial to
the use of the natural 13C labeling technique, since it works only in the first year of C3–C4 vegetation change. Nine steel collars were installed on the plot, each collar 10 cm away from a maize
shoot. The steel collars (11 cm diameter, 10 cm height) were inserted 5 cm deep into the soil to trap
CO2 from the soil. An additional nine steel collars were established on a bare fallow plot located
next to the maize plot. Both plots were fertilized with 202 kg N ha–1, equally provided as ammonium and nitrate. The plots were kept free of weeds by the pesticide dimetheanid-P (900 g
ha–1) and by manual weeding.
Sampling and analyses
On day 39 (July 16, 2004) after germination of the maize, a glass dish filled with 20 ml 1 M sodium
hydroxide (NaOH) solution was placed into every collar of the maize and bare fallow plots, and the
collars were sealed with a plastic lid. The CO2 efflux from the soil was trapped for 7 d in NaOH,
then the NaOH was collected from the traps and the glass dishes were rinsed with 20 ml deionized
water, which was mixed with the NaOH samples. The trapping and sampling procedure was repeated at day 117 of maize growth (September 22, 2004).
On the first day of each CO2-trapping period, soil samples were taken from 0 – 5 cm depth next
to the steel collars. The moist soil samples were immediately frozen until preparation for microbial
biomass analyses. We assume that there is no significant difference between microbial C and N
concentrations of field moist and frozen soil, which was approved by Stenberg et al. (1998). The
maize was harvested in mid-October, when the plants were 146 d old. The roots were carefully
washed with deionized water to remove adhering soil particles. Shoots and roots were dried at
40°C.
To estimate the total CO2 efflux, the CO2 trapped in NaOH solution was precipitated with 0.5 M
barium chloride (BaCl2) solution and then the NaOH was titrated with 0.2 M hydrochloric acid
(HCl) against phenolphthalein indicator (Zibilske, 1994). Soil microbial biomass was determined by
the chloroform fumigation extraction method (Vance et al., 1987), in which the typical extractant
concentration of 0.5 M potassium sulphate (K2SO4) solution was reduced to 0.05 M in order to increase the ratio of extracted C and N to salt to facilitate analysis by direct combustion prior to mass
spectrometer analyses. Bruulsema and Duxbury (1996) have shown that the same extraction efficiency factors (kEC and kEN) can be used in this modified method compared to the original method.
5 Partitioning of CO2 efflux from soil planted with maize
89
Aggregates of the unfrozen soil were destroyed with tweezers and roots were carefully removed
from the sample by handpicking. 10 g of soil were extracted with 40 ml of the K2SO4 solution. Another 10 g of soil were firstly fumigated with chloroform for 24 h and then extracted in the same
way. The K2SO4 and soil mixtures were shaken for 1 h on a horizontal shaker, centrifuged at 1449 ·
g for 10 min, and then filtrated through a ceramic vacuum filter. The extracts were frozen until
analyses for total carbon (C) and nitrogen (N) concentrations were done with a Dimatoc-100
TOC/TIC analyzer (Dimatec, Germany). Microbial biomass C and N concentrations were calculated from these results using a kEC value of 0.45 (Wu et al., 1990) and a kEN value of 0.54 (Brookes
et al., 1985) and are presented in percent of dry soil. The soil water content was determined in another 10 g of soil, which was dried at 105°C. These soil samples and an aliquot of the leaf or root
samples were ground with a ball mill before analysis. The total C and N content in leaves, roots,
and soil was measured with a Euro EA C/N analyzer (EuroVector, Italy).
A Thermo Finnigan MAT Delta plus Advantage isotope ratio mass spectrometer (IRMS) was
coupled to this C/N analyzer to measure δ13C values in shoots, roots, and soil. Since only solid
samples could be analyzed by the IRMS unit, the CO2 and microbial biomass samples had to be
specifically prepared. Any CO2 trapped as sodium carbonate (Na2CO3) in an aliquot of 4 ml of
NaOH was precipitated with 5 ml of 0.5 M strontium chloride (SrCl2) aqueous solution. The NaOH
solutions containing the SrCO3 precipitants were then centrifuged three times at 1449 · g for 10 min
and washed in between with deionized and degassed water to remove NaOH and to reach a pH of
7.0. After washing, the remaining water was removed from the vials and the SrCO3 was dried at
105°C. The SrCO3 was analyzed using the IRMS for δ13C values. For the microbial biomass, an
aliquot of the K2SO4 samples was pipetted directly into tin capsules and dried at 60°C prior to
IRMS analyses (according to Brant et al., 2006).
Calculations
In the root exclusion approach, root-derived CO2 and SOM decomposition were separated by measuring the total CO2 efflux from maize and bare fallow plots and by subtracting the latter from the
former. This difference between the two kinds of plots was taken to be the contribution of rootderived CO2 to the total CO2 efflux from the maize plot. The contribution of SOM decomposition
on the maize plot was considered to be equal to the total CO2 efflux from the bare fallow plot, since
no plant residues remained on the latter from the previous crop. Contributions of root-derived respiration (RDRre) and SOM decomposition (SOMDre) to the total CO2 efflux from soil planted with
maize in the root exclusion approach were calculated by the following equations:
RDRre =
C m − C bf
Cm
⋅ 100%
(5-1),
90
5 Partitioning of CO2 efflux from soil planted with maize
SOMDre =
C bf
Cm
⋅ 100%
(5-2),
where Cm and Cbf are the amounts of C from maize and bare fallow plot CO2 effluxes, respectively.
In the second – natural 13C labeling – approach, the method proposed by Kuzyakov (2004; 2005)
was used to separate RR, RMR, and SOMD. The following equations were used to calculate δ13C
values and CO2 efflux partitioning. A mass balance equation was used to determine the δ13C value
of microbial biomass (δ13CMB):
δ C MB =
13
δ 13C fum ⋅ C fum −δ 13C nf ⋅ C nf
C fum − C nf
(5-3),
where δ13Cfum and δ13Cnf are the δ13C values of the fumigated and non-fumigated samples, respectively, and Cfum and Cnf are the amounts of C in the fumigated and non-fumigated K2SO4 samples,
respectively.
In the beginning of every CO2 trapping there was a small volume of atmospheric CO2 inside the
steel collar. We considered this atmospheric CO2 in relation to the measured δ13C value by a mass
balance equation:
δ 13CCO2 =
δ 13Ctotal ⋅ Ctotal −δ 13C air ⋅ C air
Ctotal − C air
(5-4),
where δ13CCO2 is the δ13C value of soil air without atmospheric air, δ13Ctotal is the measured δ13C
value of CO2, δ13Cair is the δ13C value of ambient air (as an approximation –7.8‰ was taken from
Boutton, 1991), Ctotal is the amount of CO2-C trapped in NaOH, and Cair is the amount of C inside
the steel collar (0.08 mg C) calculated from a CO2 concentration of 345 ppm (Boutton, 1991) and
the aboveground volume of air inside the steel collar.
After calculating the δ13C of microbial biomass (by Eq. (5-3)) and the corrected δ13C of total
CO2 efflux (by Eq. (5-4)), it was possible to calculate below-ground CO2 partitioning. The development of the equations used to calculate the three sources CO2 partitioning is presented in detail
by Kuzyakov (2004; 2005). The equations for the contributions of SOM decomposition (SOMDnl)
and rhizomicrobial respiration (RMRnl) used in the 13C natural labeling (hence the subscript nl) approach are:
SOMDnl =
RMRnl =
δ 13 CCO2 − δ 13 C Rhiz
δ 13 C SOM − δ 13 C Rhiz
⋅100%
(δ 13 C MB − δ 13 C SOM ) ⋅ (δ 13 C CO2 − δ 13 C Rhiz )
(δ 13 C Rhiz − δ 13 C SOM ) ⋅ (δ 13 C MB − δ 13 C Rhiz )
(5-5),
⋅ 100% (5-6),
where δ13CCO2 is the δ13C value of the total CO2 efflux from planted soil (Eq. (5-4)), δ13CRhiz is the
δ13C value of C4 plant roots, δ13CSOM is the δ13C value of SOM from unplanted soil, and δ13CMB is
5 Partitioning of CO2 efflux from soil planted with maize
91
the δ13C value of microorganisms from planted soil (Eq. (5-3)). Having calculated these two contributions to the below-ground CO2 efflux, the remaining part would be root respiration (RRnl):
RRnl = 100% − SOMDnl − RMRnl
(5-7).
When isotopic fractionation was considered between SOM and CO2 derived from SOM and between microbial biomass and microbially derived CO2, δ13CSOM and δ13CMB were replaced in Eqs.
(5-5) and (5-6) by δ13CSOM–CO2 and δ13CMB–CO2:
δ 13 C SOM −CO2 = δ 13 C SOM + ε SOM −CO2
(5-8),
δ 13 C MB −CO2 = δ 13 C MB + ε MB −CO2
(5-9),
where εSOM–CO2 and εMB–CO2 are 13C isotopic fractionations as absolute values in ‰ between SOM
and CO2 and between microbial biomass and CO2, respectively.
A calculated δ13C value was used to determine the influence of active and inactive microbial
biomass fractions on δ13C of total microbial biomass according to Werth et al. (2006). This δ13C
value (δ13CcalcMB) was calculated by a mass balance equation using δ13C values of maize roots
(δ13CRhiz) for active (δ13Cactive) and δ13C values of SOM (δ13CSOM) from unplanted soil for inactive
(δ13Cinactive) portions of microbial biomass including the fractionation εSOM–CO2 between the substrate
and the CO2:
δ 13C calcMB =
δ 13C active ⋅ C active +δ 13C inactive ⋅ C inactive
100%
(5-10),
δ 13 C active = δ 13 C Rhiz + ε SOM −CO2
(5-11),
δ 13 C inactive = δ 13 C SOM + ε SOM −CO2
(5-12),
where Cactive and Cinactive are C proportions of active and inactive microbial biomass, respectively.
Cactive was adjusted to match measured results of below-ground CO2 partitioning (see in the results
section), and Cinactive was 100% – Cactive. The fractionation between maize rhizodeposits and CO2
was assumed to be the same as the fractionation between SOM and CO2 (compare Balesdent and
Mariotti, 1996; Boutton, 1996; Bol et al., 2003).
The total CO2 efflux (in mg) from the maize plot was split up into different sources by multiplying with the percentage contributions gained from Eqs. (5-1), (5-2), (5-5), (5-6), and (5-7). The
means of all results were calculated for maize and bare fallow plots, and standard deviations (SD)
were calculated as a variability parameter. Outliers in the CO2 trapping were excluded from the calculation of the mean CO2 efflux (and its SD), which lead to a total of six collars (instead of nine).
For calculations of mean δ13C values, we used only three replicates, since we sampled three plants
randomly. We determined the δ13C values only in the three corresponding soil and microbial biomass samples. A one-way analysis of variance (ANOVA) was used to identify differences between
total C concentrations, total N concentrations, or C/N ratios of various plant or soil pools. The
92
5 Partitioning of CO2 efflux from soil planted with maize
ANOVA was conducted pairwise between plant parts, between sampling dates of an individual plot,
or between plots for individual sampling dates. For the δ13C values, a one-way ANOVA was used
to find differences between sampling dates of an individual plot, between plots for an individual
sampling date, or between carbon pools for an individual sampling date and plot. For the latter, a
Fisher LSD test was used as post hoc test to identify differences in δ13C values of individual carbon
pool pairs. Since variances were not equal for maize plot carbon pools on the September sampling,
a Studentized maximum modulus test had to be used as post hoc test here. One-way ANOVA was
also used when the contributions to the total CO2 efflux were compared between the root exclusion
and natural 13C labeling approaches. Statistics were calculated with the SPSS 10.0 package.
Results
Carbon and nitrogen concentrations and C/N ratios
No significant difference in total C concentration was determined between shoots and roots; it averaged about 44% (Table 5-2). The shoots’ total N concentration was about twice as high as in the
roots (P < 0.01), doubling the C/N ratio in roots versus shoots (P < 0.01). The total C concentration
in the soil (Table 5-2) remained constant between the two sampling dates at 1.2% and 1.1% for the
maize and bare fallow plots, respectively. On the maize plot, the total soil N concentration declined
between spring and autumn samplings (P < 0.001). This decline significantly increased the soil C/N
ratio by 4 between the two sampling dates (P < 0.001). In July, the total C concentration in the microbial biomass was about 0.008% of soil weight on the maize plot and 0.012% of soil weight on
the bare fallow plot (Table 5-2). This between-plot difference (P < 0.05) and a nearly constant N
concentration in microbial biomass of 0.003% of soil weight on both plots led to smaller C/N ratios
on the maize versus bare fallow plot. Due to the high SD on the bare fallow plot this difference was
not significant.
5 Partitioning of CO2 efflux from soil planted with maize
93
Table 5-2: Total carbon and nitrogen concentrations and C/N ratios of shoots, roots, soil, and microbial biomass of maize and bare fallow plots (means ± SD, 2 ≤ n ≤ 9, n.d. = not determined), based on plant part or
soil dry matter. Values followed by the same first letter within columns are not significantly different (P >
0.05) between plant parts, or between soil and microbial biomass sampling dates of an individual plot. Values followed by the same second letter within columns are not significantly different (P > 0.05) between
plots for individual sampling dates.
Sampling date
in 2004
C
[% of dry matter]
maize
shoots
29.09.
42.16
a
roots
29.09.
45.27
a
14.07.
1.19
b,d
29.09.
1.09
b,e
14.07.
1.09
c,d
29.09.
1.13
c,e
14.07.
0.008
f,h
29.09.
0.011
f,j
14.07.
0.012
g,i
0.012
g,j
N
[% of dry matter]
± 1.54
± 8.70
2.43
a
1.26
b
± 0.05
± 0.17
0.27
c,f
0.13
d,h
± 0.09
± 0.05
0.16
e,g
0.14
e,h
± 0.002
± 0.006
0.004
i,j
0.003
i
± 0.004
± 0.002
0.003
j
C/N
± 0.19
± 0.18
17.4
a
36.0
b
± 0.01
± 0.01
4.5
c,f
8.3
d,h
± 0.01
± 0.01
7.0
e,g
8.0
e,h
± 0.001
± 0.002
3.1
i,j
6.9
i
± 0.002
n.d.
7.4
j
± 0.8
± 5.3
soil
maize plot
bare fallow plot
± 0.4
± 0.4
± 0.9
± 0.1
microbial biomass
maize plot
bare fallow plot
29.09.
± 0.8
± 4.2
± 4.2
n.d.
δ13C values
In none of three carbon pools (i.e. SOM, microbial biomass, and soil-derived CO2) on the maize
plot did δ13C values change significantly between sampling dates Fig. 5-1a). On the September
sampling date, the δ13C value of microbial biomass on the maize plot was significantly more positive (by 1.5‰) than that of SOM (–25.5‰, P < 0.05). On both sampling dates, CO2 from soil respiration was significantly enriched in 13C (by about 5.0‰) compared to microbial biomass (P < 0.05).
The δ13C of maize roots (–12.2‰) was more positive than the δ13C of CO2 (–18.6‰, P < 0.05).
Maize roots were significantly more enriched in 13C than leaves (–13.2‰, P < 0.01).
On the bare fallow plot, the δ13C value of CO2 from soil respiration was more positive in September than in July (by 3.8‰, P < 0.001, Fig. 5-1b). Only the δ13C values of SOM and CO2 differed
significantly (by 1.6‰) on the first sampling (P < 0.01). The difference between microbial biomass
and CO2 was only 0.7‰, which was not significant. On the second sampling, all δ13C values were
94
5 Partitioning of CO2 efflux from soil planted with maize
significantly different. SOM was depleted by 1.6‰ (P < 0.01) and CO2 was enriched by 3.7‰ (P <
0.001) compared to microbial biomass (δ13C = –24.0).
The δ13C values of SOM and microbial biomass did not differ between the two plots. However,
the CO2 efflux from soil was significantly enriched in 13C on the maize versus bare fallow plot (P <
0.05).
Fig. 5-1: δ13C values of carbon pools from (a) maize and (b) bare fallow plots. Carbon pools are maize leaves
({), maize roots (…), soil organic matter (‘), total CO2 efflux (U), and microbial biomass (¯); error bars
show standard deviation (n = 3). Values followed by the same letter within one sampling month are not significantly different (P > 0.05).
CO2 efflux partitioning
The comparison of total CO2 effluxes from maize and bare fallow plots revealed that the value from
the maize plot was derived 71 to 78% from SOM decomposition and 22 to 29% from roots (Fig.
5-2a). This increase of the SOMD contribution was not significant between the two sampling dates.
In contrast to the root exclusion approach, the contribution of SOMD to the total CO2 efflux estimated by the natural 13C labeling technique decreased between sampling dates from 56 to 48%
(Fig. 5-2a). Between dates, the contributions of RR and RMR increased from 37 to 40% and from 7
to 12%, respectively. None of these changes between sampling dates were significant. 13C fractionations between SOM and CO2 as well as between microbial biomass and CO2 were obtained from
the respective δ13C values of the bare fallow plot (Fig. 5-1b) and were used for further calculations.
The fractionations between microbial biomass and CO2 (εMB–CO2) were 1.0 and 4.0‰ in July and
September, respectively. At both dates, fractionations between SOM and CO2 (εSOM–CO2) were 1.0‰
higher than between microbial biomass and CO2. Using these values, we calculated below-ground
CO2 partitioning for three CO2 sources (Fig. 5-2a: right column in each month). In September, we
found significant differences (P < 0.05) between calculations with and without 13C fractionation
(Fig. 5-2a) for SOMD and for the total root-derived respiration (RDR).
5 Partitioning of CO2 efflux from soil planted with maize
95
In the comparison of the root exclusion and the natural 13C labeling approach (including 13C fractionation), root-derived CO2 (and accordingly SOM-derived CO2) was the same (P > 0.05) between
approaches in both months, July and September (Fig. 5-2a). Translated into g C m–2, root-derived
CO2-C was only 0.8 g m–2 more in the July sampling week and 0.1 g m–2 less in the September
sampling week in the natural 13C labeling versus the root exclusion approach (Fig. 5-2b).
Fig. 5-2: Carbon dioxide efflux partitioning of the maize plot calculated by the root exclusion (n = 6) and
natural 13C labeling approaches (n = 3) without and with 13C fractionation between microbial biomass and
CO2 (1‰ in July, 4‰ in September) and between SOM and CO2 (2‰ in July, 5‰ in September). Results
are shown in (a) percentage and (b) absolute values after one week of CO2 trapping. Contributions are root
respiration (no shading), rhizomicrobial respiration (hatched shading), total root-derived respiration (black
shading), and soil organic matter (SOM) decomposition (dotted shading). Error bars show standard deviation.
96
5 Partitioning of CO2 efflux from soil planted with maize
Influence of active microbial biomass on below-ground CO2 partitioning
We have redrawn the CO2 efflux partitioning by the natural 13C labeling technique including fractionations between SOM and CO2 and between the microbial biomass and CO2 from Fig. 5-2a
(right column in each month) into Fig. 5-3 (left column in each month). In order to simulate the
influence of active and inactive fractions of the microbial biomass on CO2 partitioning, we used
calculated δ13C values for the microbial biomass that considered both fractions (see Eq. (5-10)).
Percentages of these fractions in Eq. (5-10) were adjusted to match the CO2 partitioning results obtained in this study (Fig. 5-3: middle column in each month) and literature results (Fig. 5-3: right
column in each month). Values of δ13C for roots (–12.2‰) and for SOM (–25.8‰ in July, –25.6‰
in September) were used to represent δ13C values for active and inactive microbial biomass fractions, respectively. Fractionations of 2‰ in July and 5‰ in September between the substrates (i.e.
roots or SOM) and the CO2 were included in the calculations.
Active portions of the total microbial biomass, which feed on maize rhizodeposits, of about 5%
in July and about 6% in September (Fig. 5-3: middle column in each month) were determined to
reflect the results observed in this study (Fig. 5-3: left column in each month). Hypothetical active
portions of the total microbial biomass of 18% in July and 8% in September (Fig. 5-3: right column
in each month), however, would have been necessary to yield a 50% contribution each for RR and
Fig. 5-3: Influence of the active portion (in percentage on the x-axis) of total microbial biomass on belowground CO2 partitioning. In the middle column in each month the active microbial biomass was adjusted to
achieve calculated CO2 partitioning results of this present study (left column, including isotopic fractionation
according to Fig. 5-2). In the right column in each month the active portion of microbial biomass was adjusted to achieve CO2 partitioning results of literature studies (Table 5-1). Patterns are: contributions of root
respiration (no shading), rhizomicrobial respiration (hatched shading), and SOM decomposition (dotted
shading) to total CO2 efflux from a C3 soil planted with maize; error bars show standard deviation (n = 3).
5 Partitioning of CO2 efflux from soil planted with maize
97
RMR related to total RDR as reported in various studies (Table 5-1).
Discussion
Advantages and disadvantages of the two approaches
The two approaches yielded different results of CO2 efflux partitioning under field conditions, when
any fractionation processes were disregarded. In the root exclusion approach, about 20 to 30% of
the total efflux were root-derived, as opposed to about 50% estimated by the natural 13C labeling
technique (Fig. 5-2a: first two columns in each month). There are two advantages of the root exclusion approach: (1) it uses the absolute CO2 efflux and not the isotopic ratios, and (2) there were no
uncertainties as by 13C fractionation. These advantages are offset by several disadvantages: (1) the
water regime and temperature balance in planted soil may differ considerably from those in unplanted soil (Fisher and Gosz, 1986; Rochette et al., 1999b; Ross et al., 2001; Jones et al., 2004),
(2) the cycling of nutrients such as N, which affects the C cycle, also varies between vegetated and
non-vegetated soil (Rochette et al., 1999b; Hinsinger et al., 2005), and (3) the decomposition of
SOM and other plant residues is dependent on both the physical effects (water regime and temperature balance) of vegetation on soil and the direct biological effects of living roots (reviewed by
Dormaar, 1990; Kuzyakov, 2002a; Kuzyakov, 2002b; Paterson, 2003; Cheng and Kuzyakov, 2005).
Consequently, exudation from maize roots could either increase (or decrease) SOM decomposition
by priming effects (Kuzyakov et al., 2000). Since we calculated SOM decomposition only from the
bare fallow plot, priming effects cannot be considered here. Accordingly, the real contribution of
SOM decomposition on the maize plot could be slightly higher (or lower) than estimated by the root
exclusion approach. However, we neglected influences of priming effects on SOMD, since their
contribution to the total CO2 efflux from agricultural soils is usually low. This lack of priming effects is related to the fertilization of the soil and the resulting low exudation of crops and low stimulation of the microbial biomass to mineralize additional SOM (Cheng and Coleman, 1990; Paterson
and Sim, 1999; 2000). The microbial biomass on a non-fertilized neighboring grassland plot (data
not shown) was accordingly two to three times bigger than on the maize plot. All three disadvantages could be avoided when using the natural 13C labeling technique instead of the root exclusion
method.
Assumptions of the natural 13C labeling approach
The natural 13C labeling method involves two assumptions concerning
root and microbial respiration (Kuzyakov, 2004; 2005):
13
C isotopic effects during
1) The δ13C isotope signature of root-derived CO2 is the same as the δ13C value of the roots.
98
5 Partitioning of CO2 efflux from soil planted with maize
2) The δ13C isotope signature of CO2 respired by microorganisms corresponds to the δ13C value
of microbial biomass.
The first assumption has been accepted by Werth et al. (2006), since 13C depletion of root-derived
CO2 ranges from 0.2‰ in sand (Cheng, 1996) to 0.7‰ in nutrient rich hydroculture (Werth and
Kuzyakov, 2006) compared to the δ13C value of the roots. The second assumption, however, cannot
be accepted and hence 13C fractionations between microbial biomass and microbial CO2 have to be
considered (Werth et al., 2006) – 1 and 4‰ in July and September, respectively. Additionally, we
had to consider isotopic fractionations between SOM and CO2, which were 2‰ in July and 5‰ in
September. We call these differences ‘fractionations’, although they are not only fractionations in a
kinetic sense (the heavier 13C isotope reacts more slowly than the lighter 12C isotope). They also
include preferential utilization of substrates with different biological availability having not identical δ13C values. The first fractionation step leading to a 13C-enriched microbial biomass compared
to SOM can be explained by isotope discrimination during biosynthesis of new microbial biomass
(Potthoff et al., 2003). Compared to SOM, water-soluble organic compounds with a heavier isotopic composition are preferentially used by soil microorganisms (Pelz et al., 2005). The second
fractionation step results in more 13C-enriched microbial CO2 compared to the microbial biomass
and the substrate. Usually, CO2 from microbial respiration is 13C-depleted compared to the feeding
substrate (Blair et al., 1985; Mary et al., 1992; Potthoff et al., 2003). Opposite results, i.e. an 13C
enrichment of the CO2 compared to the source, can be explained by a selective use of 13C-enriched
SOM compounds by microorganisms (Werth et al., 2006). This selection was more pronounced
than the 13C depletion effect of the metabolism itself (Šantrůčková et al., 2000), resulting in 13Cenriched CO2. Total fractionation between SOM and CO2 and between microbial biomass and CO2
including kinetic and biological processes is important, when using the method proposed by
Kuzyakov (2004; 2005). Fractionations will also be of importance in studies that use similar 13C
natural abundance methods including mass balance equations with soil or microbial biomass substitutes to soil or microbial CO2.
A 13C enrichment of the trapped CO2 used for further calculations due to the mixing with atmospheric CO2 can be excluded, since we subtracted the δ13C of that CO2 inside the sampling collar
from the δ13C of the total CO2 sampled (Eq. (5-4)). By using the same natural 13C labeling approach
in our lab experiment (Werth et al., 2006) like in this field experiment, we determined fractionations of 2 and 5‰ between microbial biomass and CO2 and between SOM and CO2, respectively. In
the former experiment, we trapped almost 100% of the CO2, since we used a small closed system
with continuous air circulation. The fractionations determined in the field experiment can be approved by the results of the lab experiment, since they were in the same range. In order to prevent a
possible fractionation due to trapping and precipitation of CO2 in alkali, direct gas sampling (e.g.
with a syringe) and mass spectrometry analysis should be preferred in future studies. However, the
mixing of the chamber air with atmospheric CO2 cannot be excluded by direct gas sampling. This
can only be reduced by taking subsurface gas samples (Kammann et al., 2001; Schneckenberger
and Kuzyakov, 2006), but then the focus will no longer be on CO2 efflux, but rather on CO2 compo-
5 Partitioning of CO2 efflux from soil planted with maize
99
sition in a certain soil depth. Furthermore, even with this approach, isotopic fractionation due to
CO2 diffusion through the soil cannot be excluded.
CO2 efflux partitioning by the root exclusion and natural 13C labeling approaches
Consideration of these fractionations led to CO2 partitioning results that were much closer to the
root exclusion method than those calculated without considering 13C fractionations (Fig. 5-2a). Several other studies reported 13C fractionations in this range (1 – 5‰) (Rochette et al., 1999a; Šantrůčková et al., 2000; Formánek and Ambus, 2004). In September, the contributions of root- and
SOM-derived CO2 to the total CO2 efflux were 22% and 78% (Fig. 5-2a), respectively. These contributions calculated by the 13C labeling approach (taking fractionation into account) match the root
exclusion results very well (Fig. 5-2a). Hence, the below-ground CO2 partitioning results of the
former could be adjusted to the results of the latter by considering 13C isotopic fractionation processes.
The general dominance of the SOMD contribution to the total CO2 efflux from soil in both approaches can be explained by the CO2 sampling method: the steel collars for CO2 trapping were too
small to encompass the whole plant with its rhizosphere. To avoid injuring the upper roots, the collars were placed about 10 cm away from the shoots. However, since the steel collars were inserted
about 5 cm deep into the soil, we cannot absolutely exclude severance of the roots. Either a possible
severance of the roots, the long distance of the CO2 trap to the shoot, or a combination of both effects could have led to underestimated contributions of root-derived CO2. On the contrary, the
SOMD contribution to the total CO2 efflux could have been overestimated due to contributions of
root-free soil. Hence, further studies should prefer to use CO2 trapping collars, which encompass
the complete rhizosphere. In other studies, the average contribution of root-derived CO2 from various grass species rhizospheres was also lower in the field than in the laboratory (Table 5-1). This
difference can be explained by the different environmental influences (soil moisture, temperature,
etc.) between experiments under controlled and field conditions.
In contrast to the present field study, where SOMD was the dominant CO2 source, we calculated
an average of 85% root respiration by the natural 13C labeling technique under controlled conditions
(Werth et al., 2006). These high RR contributions reflect a strongly artificial root-to-soil ratio under
controlled conditions. Rhizomicrobial respiration and SOM decomposition were both remarkably
low under controlled conditions when compared to other studies. Opposite results from our field
and laboratory study for RR and SOMD, and the strong influence of isotopic fractionation on CO2
partitioning results, show that the natural 13C labeling technique still remains to be adjusted to these
influences.
In the calculations for CO2 efflux partitioning (Fig. 5-2a) without considering isotopic fractionation and in July, when a low fractionation between SOM and CO2 and between microbial biomass
and CO2 was considered, 4 to 12% RMR were remarkably low compared to the mean of 24% of
100
5 Partitioning of CO2 efflux from soil planted with maize
other studies (Kelting et al., 1998; Crow and Wieder, 2005; Chen et al., 2006). Literature results
have shown nearly equal contributions of RR and RMR to root-derived CO2 (see Table 5-1). Even
the smallest contributions of RMR determined by combining root exclusion and component integration methods on a maize and spring barley (Hordeum vulgare L.) field amounted about 40% of total
root-derived CO2 (see Table 1, Larionova et al., 2006) and were between double to four times as
high as in our field experiment. These differences can also be explained by the position of the steel
collars as discussed for SOMD and/or by low activity of microorganisms.
Influence of active microbial biomass on below-ground CO2 partitioning
Activity versus dormancy of soil microorganisms is crucial when calculating RMR using the natural
13
C labeling technique, since the method of Kuzyakov (2004; 2005) uses the δ13C value of microbial biomass itself as a substitute for the δ13C value of microbial CO2 (the sum of RMR and
SOMD). It is not possible to separate microbial CO2 directly. In order to determine the active fraction of microbial biomass contributing to the δ13C of microbial biomass, we used a calculated δ13C
value (Eq. (5-10)) by mass balance of C3 and C4 source contributions. A maximum of a 6% active
portion of the total microbial biomass contributing to the microbial CO2 efflux was calculated in
this field study (Fig. 5-3). This is not an estimate for the ‘real’ active microbial biomass in the field,
but rather a determination of the hypothetic active microbial biomass when using the natural 13C
labelling technique according to Kuzyakov (2004; 2005), including the second assumption of equal
δ13C values of microbial biomass extracts and microbial CO2. In a laboratory study with maize
grown on the same soil as in the field study, we showed that – without fractionation between the
substrate and the CO2 in the calculation of the δ13C value of microbial CO2 – about 37% of the microbial biomass in the rhizosphere was active (Werth et al., 2006). Considering also a fractionation
in the microbial substrate respiration of 5‰ when calculating the δ13C of microbial respiration (according to Eq. (5-10)) would reduce the active microbial biomass in the laboratory study to 9%.
This still higher activity compared to the field study can be explained by the controlled conditions
(water content, temperature etc.), the small size of the plant growing containers, and the resulting
sampling of rhizosphere soil closely related to the plants. On the contrary, the lower activity in the
field can be explained by sub-optimal weather conditions, the longer distance of the soil sampling
location to the center of the root system, and the resulting stronger contribution of SOM feeding or
passive microorganisms to the total microbial biomass. In both experiments, substituting the δ13C
value of microbial CO2 with the δ13C value of microbial biomass led to the following discrepancy:
microbial biomass obtained by fumigation-extraction procedure was dominated by dormant microorganisms that had fed earlier mainly on a C3 source, whereas δ13C of microbial CO2 mainly derived from active microorganisms feeding primarily on a C4 source. Increasing the proportion of
active microbial biomass in the mass balance equation for calculating RMR raised its contribution
to the total CO2 efflux (Fig. 5-3). In the present field study, the δ13C values of microbial biomass
from maize and bare fallow plots were the same (Fig. 5-1). This equality indicates that only a very
small active portion of microorganisms utilized maize rhizodeposits and that the δ13C of microbial
5 Partitioning of CO2 efflux from soil planted with maize
101
biomass was mainly influenced by dormant or SOM-feeding microorganisms. Consequently, the
δ13C of microbial biomass is not an appropriate substitute for the δ13C of microbial respiration. Besides isotopic fractionation between microbial biomass and CO2, the discrepancy between active
microbial contributions in the biomass itself and in the CO2 enforces us to disprove the assumption
of equal δ13C values of microbial biomass and CO2 when using the natural 13C labeling technique.
Thus, there is no single source of CO2 contribution from the microbial community, but rather a
small contribution from the large fraction of dormant microorganisms and a large contribution from
the small active fraction. Consequently, a new method has to be found to determine the activity of
the microbial biomass independently from our mass balance calculations and then to calculate the
δ13C of microbial respiration with the proportion obtained from that method in mass balance Eq. (510). Subsequent calculations of CO2 partitioning could lead to results much closer to former experiments (Table 5-1).
Seasonal effect on below-ground CO2 partitioning
The δ13C values of plant parts were only measured on the autumn sampling. In our 40 d experiment
under controlled conditions (Werth et al., 2006), we did not find any differences in δ13C values of
plant parts from different sampling dates. However, during a whole plant growth period, like in this
field experiment, δ13C values of different plant tissues could change with plant growth. Giesemann
et al. (2006) have reported no significant changes in shoots δ13C values during the growth period of
winter barley (Hordeum vulgare L.) and winter wheat (Triticum aestivum L.). However, they found
a depletion of 1.5‰ in the shoots of sugar beet (Beta vulgaris L.) during plant growth. According to
this, 13C-enriched roots in our maize plants in July would have a significant impact on the results of
CO2 efflux partitioning. Consequently, in future studies δ13C values in plant tissues should be
measured on each sampling date.
On both plots, the total CO2 efflux in September was half as high as in July (Fig. 5-2b). This
typical decline in autumn has been reported by several studies under comparable climate conditions
(Rochette et al., 1999b; Kutsch et al., 2001; Amos et al., 2005; Han et al., 2007). This decline reflects lower air and soil temperatures in September and thus lower total respiration of plant roots
and soil microorganisms. Low root respiration rates late in the season can be explained by a decrease of the mass ratio of respiring fine roots to structural, coarse roots as roots age (Lipp and Andersen, 2003). Whereas absolute soil respiration declined in our field experiment (Fig. 5-2b), the
contribution of SOMD to the total CO2 efflux increased from 65 to 78% whilst the contribution of
RMR increased from 4 to 8% (Fig. 5-2a). Maize plants were already in the senescence stage in September, i.e. parts of the roots were dying and became decomposed by rhizosphere microorganisms.
Consequently, the contribution of SOMD was higher in September than in July due to less rootderived CO2; the contribution of RMR increased due to reduced root respiration and decomposition
of dead root cells by rhizosphere microorganisms. Total RDR in September, however, may have
102
5 Partitioning of CO2 efflux from soil planted with maize
been underestimated due to higher δ13C values of CO2 from SOM decomposition than in July. Similar studies have reported an increased δ13C value of CO2 from bare soil respiration late in the season
(Rochette and Flanagan, 1997; Rochette et al., 1999b). One interpretation is that this late-season
δ13C increase reflects a reduced soil respiration rate. The change in weather conditions in early autumn may also have led to a convective transfer of CO2 downward from the aboveground atmosphere into the soil when the temperature is cooler at the soil surface than below (Rochette et al.,
1999b). Consequently, the higher δ13C value of CO2 from bare soil in autumn (Fig. 5-1b) might be
less a result of fractionation than of weather-induced mixing of soil CO2 and atmospheric CO2. The
fractionations of 4‰ between microbial biomass and CO2 – and of 5‰ between SOM and CO2 – in
September would then be overestimations. Lower fractionations would lead to a higher contribution
of root-derived CO2 to the total CO2 efflux from the maize plot than shown in Fig. 5-2a. This calls
for further clarification as to whether the higher δ13C value of CO2 from bare soil respiration in autumn reflects fractionation or changing weather conditions.
Conclusions
The natural 13C labeling technique allowed partitioning of the total CO2 efflux from soil into three
sources: SOM decomposition, root respiration and rhizomicrobial respiration. Positioning the CO2
trapping collars between the maize plants underestimated the portion of root-derived CO2 and overestimated the contribution of SOM decomposition to the total CO2 efflux from soil. The consideration of 13C fractionation by respiration of microbial biomass and SOM decomposition was crucial
for acceptable results on CO2 efflux partitioning of a maize plot on an arable loamy Haplic Luvisol.
At the end of the growth period, SOM decomposition contributed 78% to the total CO2 efflux,
whereas RR and RMR amounted to 14% and 8%, respectively. The root exclusion approach confirmed this partitioning by the natural 13C labeling method with similar results of SOM- and rootderived CO2. Nevertheless, there are uncertainties by both approaches like different environmental
or nutrient conditions in planted and unplanted soils in the root exclusion approach or isotopic fractionations in the natural 13C labeling approach, which have to be addressed in further investigations.
Acknowledgements
The German Research Foundation (DFG) kindly supported this work. The authors would like to
thank H. Stelz for the field plot maintenance and Dr. W. Armbruster and E. Dachtler for the IRMS
analyses.
References
Amos, B., Arkebauer, T.J., Doran, J.W., 2005. Soil surface fluxes of greenhouse gases in an irrigated maize-based agroecosystem. Soil Science Society of America Journal 69, 387-395.
5 Partitioning of CO2 efflux from soil planted with maize
103
Andrews, J.A., Harrison, K.G., Matamala, R., Schlesinger, W.H., 1999. Separation of root respiration from total soil respiration using carbon-13 labeling during free-air carbon dioxide
enrichment (FACE). Soil Science Society of America Journal 63, 1429-1435.
Balesdent, J., Mariotti, A., 1996. Measurement of soil organic matter turnover using 13C natural
abundance. In: Boutton, T.W., Yamasaki, S.I. (Eds.), Mass Spectrometry of Soils, Marcel
Dekker, New York, USA, pp. 83-111.
Blair, N., Leu, A., Munoz, E., Olsen, J., Kwong, E., Des Marais, D., 1985. Carbon isotope fractionation in heterotrophic microbial metabolism. Applied and Environmental Microbiology
50, 996-1001.
Bol, R., Moering, J., Kuzyakov, Y., Amelung, W., 2003. Quantification of priming and CO2 respiration sources following slurry-C incorporation into two grassland soils with different C
content. Rapid Communications in Mass Spectrometry 17, 1-6.
Boutton, T.W., 1991. Stable carbon isotope ratios of natural materials: II. atmospheric, terrestrial,
marine, and freshwater environments. In: Coleman, D.C., Fry, B. (Eds.), Carbon Isotope
Techniques. Isotopic Techniques in Plant, Soil, and Aquatic Biology, Academic Press, Inc.,
San Diego, pp. 173-185.
Boutton, T.W., 1996. Stable carbon isotope ratios of soil organic matter and their use as indicators
of vegetation and climate change. In: Boutton, T.W., Yamasaki, S.I. (Eds.), Mass
Spectrometry of Soils, Marcel Dekker, New York, pp. 47-82.
Brant, J.B., Sulzman, E.W., Myrold, D.D., 2006. Microbial community utilization of added carbon substrates in response to long-term carbon input manipulation. Soil Biology & Biochemistry 38, 2219-2232.
Brookes, P.C., Landman, A., Pruden, G., Jenkinson, D.S., 1985. Chloroform fumigation and the
release of soil nitrogen: a rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biology & Biochemistry 17, 837-842.
Brumme, R., 1995. Mechanisms of carbon and nutrient release and retention in beech forest gaps.
Plant and Soil 168-169, 593-600.
Bruulsema, T.W., Duxbury, J.M., 1996. Simultaneous measurement of soil microbial nitrogen,
carbon, and carbon isotope ratio. Soil Science Society of America Journal 60, 1787-1791.
Chen, C.R., Condron, L.M., Xu, Z.H., Davis, M.R., Sherlock, R.R., 2006. Root, rhizosphere and
root-free respiration in soils under grassland and forest plants. European Journal of Soil Science 57, 58-66.
Cheng, W., 1996. Measurement of rhizosphere respiration and organic matter decomposition using
natural 13C. Plant and Soil 183, 263-268.
Cheng, W., Coleman, D.C., 1990. Effect of living roots on soil organic matter decomposition. Soil
Biology & Biochemistry 22, 781-787.
Cheng, W., Kuzyakov, Y., 2005. Root effects on soil organic matter decomposition. In: Wright, S.,
Zobel, R. (Eds.), Roots and Soil Management: Interactions between Roots and the Soil.
Agronomy Monograph, No. 48, American Society of Agronomy, Madison, pp. 119-143.
Cheng, W., Coleman, D.C., Carroll, C.R., Hoffman, C.A., 1993. In situ measurement of root
respiration and soluble C concentrations in the rhizosphere. Soil Biology & Biochemistry 25,
1189-1196.
Craine, J.M., Wedin, D.A., Chapin, F.S.I., 1999. Predominance of ecophysiological controls on
soil CO2 flux in a Minnesota grassland. Plant and Soil 207, 77-86.
Crow, S.E., Wieder, R.K., 2005. Sources of CO2 emission from a northern peatland: Root respiration, exudation, and decomposition. Ecology 86, 1825-1834.
104
5 Partitioning of CO2 efflux from soil planted with maize
Dormaar, J.F., 1990. Effect of active roots on the decomposition of soil organic materials. Biology
and Fertility of Soils 10, 121-126.
Fisher, F.M., Gosz, J.R., 1986. Effects of trenching on soil processes and properties in a New
Mexico mixed-conifer forest. Biology and Fertility of Soils 2, 35-42.
Formánek, P., Ambus, P., 2004. Assessing the use of δ13C natural abundance in separation of root
and microbial respiration in a Danish beech (Fagus sylvatica L.) forest. Rapid Communications in Mass Spectrometry 18, 1-6.
Giesemann, A., Manderscheid, R., Burkart, S., Weigel, H.-J., 2006. Carbon (C) isotopic composition of agricultural crops cultivated in a crop rotation system under FACE (Free Air Carbon Dioxide Enrichment) conditions. In: 29th Annual Meeting of the German Association for
Stable Isotope Research (GASIR), 4 to 6 October 2006, Freiberg, Germany, 2006. Ed.
GASIR. p. 37.
Gloser, J., Tesarová, M., 1978. Litter, soil, and root respiration measurement. An improved compartmental analysis method. Pedobiologia 18, 76-81.
Han, G., Zhou, G., Xu, Z., Yang, Y., Liu, J., Shi, K., 2007. Biotic and abiotic factors controlling
the spatial and temporal variation of soil respiration in an agricultural ecosystem. Soil Biology & Biochemistry 39, 418-425.
Hanson, P.J., Edwards, N.T., Garten, C.T., Andrews, J.A., 2000. Separating root and soil microbial contributions to soil respiration: A review of methods and observations. Biogeochemistry 48, 115-146.
Hinsinger, P., Gobran, G.R., Gregory, P.J., Wenzel, W.W., 2005. Rhizosphere geometry and
heterogeneity arising from root-mediated physical and chemical processes. New Phytologist
168, 293-303.
Johansson, G., 1992. Release of organic C from growing roots of meadow fescue (Festuca pratensis L.). Soil Biology & Biochemistry 24, 427-433.
Jones, D.L., Hodge, A., Kuzyakov, Y., 2004. Plant and mycorrhizal regulation of rhizodeposition.
New Phytologist 163, 459-480.
Kammann, C., Grünhage, L., Jäger, H.-J., 2001. A new sampling technique to monitor concentrations of CH4, N2O and CO2 in air at well-defined depths in soils with varied water potential. European Journal of Soil Science 52, 297-303.
Kelting, D.L., Burger, J.A., Edwards, G.S., 1998. Estimating root respiration, microbial respiration in the rhizosphere, and root-free soil respiration in forest soils. Soil Biology & Biochemistry 30, 961-968.
Killham, K., Yeomans, C., 2001. Rhizosphere carbon flow measurement and implications: from
isotopes to reporter genes. Plant and Soil 232, 91-96.
Kutsch, W.L., Staack, A., Wotzel, J., Middelhoff, U., Kappen, L., 2001. Field measurements of
root respiration and total soil respiration in an alder forest. New Phytologist 150, 157-168.
Kuzyakov, Y., 2002a. Review: Factors affecting rhizosphere priming effects. Journal of Plant Nutrition and Soil Science 165, 382-396.
Kuzyakov, Y., 2002b. Separating microbial respiration of exudates from root respiration in nonsterile soils: a comparison of four methods. Soil Biology & Biochemistry 34, 1621-1631.
Kuzyakov, Y., 2004. Separation of root and rhizomicrobial respiration by natural 13C abundance:
theoretical approach, advantages, and difficulties. Eurasian Soil Science 37, S79-S84.
Kuzyakov, Y., 2005. Theoretical background for partitioning of root and rhizomicrobial respiration
by δ13C of microbial biomass. European Journal of Soil Biology 41, 1-9.
Kuzyakov, Y., 2006. Sources of CO2 efflux from soil and review of partitioning methods. Soil Biology & Biochemistry 38, 425-448.
5 Partitioning of CO2 efflux from soil planted with maize
105
Kuzyakov, Y., Cheng, W., 2001. Photosynthesis controls of rhizosphere respiration and organic
matter decomposition. Soil Biology & Biochemistry 33, 1915-1925.
Kuzyakov, Y., Domanski, G., 2002. Model for rhizodeposition and CO2 efflux from planted soil
and its validation by 14C pulse labelling of ryegrass. Plant and Soil 239, 87-102.
Kuzyakov, Y., Larionova, A.A., 2005. Root and rhizomicrobial respiration: A review of approaches to estimate respiration by autotrophic and heterotrophic organisms in soil. Journal
of Plant Nutrition and Soil Science 168, 503-520.
Kuzyakov, Y., Kretzschmar, A., Stahr, K., 1999. Contribution of Lolium perenne rhizodeposition
to carbon turnover of pasture soil. Plant and Soil 213, 127-136.
Kuzyakov, Y., Friedel, J.K., Stahr, K., 2000. Review of mechanisms and quantification of priming effects. Soil Biology & Biochemistry 32, 1485-1498.
Kuzyakov, Y., Ehrensberger, H., Stahr, K., 2001. Carbon partitioning and below-ground translocation by Lolium perenne. Soil Biology & Biochemistry 33, 61-74.
Larionova, A.A., Sapronov, D., Lopes de Gerenju, V.O., Kuznetsova, L.G., Kudeyarov, V.N.,
2006. Contribution of plant root respiration to the CO2 emission from soil. Eurasian Soil
Science 39, 1127-1135.
Lipp, C.C., Andersen, C.P., 2003. Role of carbohydrate supply in white and brown root respiration of ponderosa pine. New Phytologist 160, 523-531.
Mary, B., Mariotti, A., Morel, J.L., 1992. Use of 13C variations at natural abundance for studying
the biodegradation of root mucilage, roots and glucose in soil. Soil Biology & Biochemistry
24, 1065-1072.
Midwood, A.J., Gebbing, T., Wendler, R., Sommerkorn, M., Hunt, J.E., Millard, P., 2006.
Collection and storage of CO2 for 13C analysis: an application to separate soil CO2 efflux
into root- and soil-derived components. Rapid Communications in Mass Spectrometry 20,
3379-3384.
Paterson, E., 2003. Importance of rhizodeposition in the coupling of plant and microbial productivity. European Journal of Soil Science 54, 741-750.
Paterson, E., Sim, A., 1999. Rhizodeposition and C-partitioning of Lolium perenne in axenic culture affected by nitrogen supply and defoliation. Plant and Soil 216, 155-164.
Paterson, E., Sim, A., 2000. Effect of nitrogen supply and defoliation on loss of organic compounds from roots of Festuca rubra. Journal of Experimental Botany 51, 1449-1457.
Pelz, O., Abraham, W.-R., Saurer, M., Siegwolf, R., Zeyer, J., 2005. Microbial assimilation of
plant-derived carbon in soil traced by isotope analysis. Biology and Fertility of Soils V41,
153-162.
Potthoff, M., Loftfield, N., Buegger, F., Wick, B., John, B., Jörgensen, R.G., Flessa, H., 2003.
The determination of δ13C in soil microbial biomass using fumigation-extraction. Soil Biology & Biochemistry 35, 947-954.
Qian, J.H., Doran, J.W., Walters, D.T., 1997. Maize plant contributions to root zone available
carbon and microbial transformations of nitrogen. Soil Biology & Biochemistry 29, 14511462.
Raich, J.W., Mora, G., 2005. Estimating root plus rhizosphere contributions to soil respiration in
annual croplands. Soil Science Society of America Journal 69, 634-639.
Robinson, D., Scrimgeour, C.M., 1995. The contribution of plant C to soil CO2 measured using
δ13C. Soil Biology & Biochemistry 27, 1653-1656.
Rochette, P., Flanagan, L.B., 1997. Quantifying rhizosphere respiration in a corn crop under field
conditions. Soil Science Society of America Journal 61, 466-474.
106
5 Partitioning of CO2 efflux from soil planted with maize
Rochette, P., Angers, D.A., Flanagan, L.B., 1999a. Maize residue decomposition measurement
using soil surface carbon dioxide fluxes and natural abundance of carbon-13. Soil Science
Society of America Journal 63, 1385-1396.
Rochette, P., Flanagan, L.B., Gregorich, E.G., 1999b. Separating soil respiration into plant and
soil components using analyses of the natural abundance of carbon-13. Soil Science Society
of America Journal 63, 1207-1213.
Ross, D.J., Scott, N.A., Tate, K.R., Rodda, N.J., Townsend, J.A., 2001. Root effects on soil carbon and nitrogen cycling in a Pinus radiata D. Don plantation on a coastal sand. Australian
Journal of Soil Research 39, 1027-1039.
Šantrůčková, H., Bird, M.I., Frouz, J., Šustr, V., Tajovský, K., 2000. Natural abundance of 13C
in leaf litter as related to feeding activity of soil invertebrates and microbial mineralisation.
Soil Biology & Biochemistry 32, 1793-1797.
Sapronov, D., Kuzyakov, Y., 2004. Separation of root and microbial respiration: comparison of
three methods. In: Eurosoil 2004 – Abstracts, 4 - 12 September 2004, Freiburg/Br., Germany, 2004. Ed. University of Freiburg. p. 354.
Schneckenberger, K., Kuzyakov, Y., 2006. Quantifizierung der Quellen der CO2-Flüsse aus dem
Boden unter Miscanthus x giganteus anhand der natürlichen 13C-Abundanz. In: 29th Annual
Meeting of the German Association for Stable Isotope Research (GASIR), 4 to 6 October
2006, Freiberg, Germany, 2006. Ed. GASIR. p. 40.
Stenberg, B., Johansson, M., Pell, M., Sjödahl-Svensson, K., Stenström, J., Torstensson, L.,
1998. Microbial biomass and activities in soil as affected by frozen and cold storage. Soil
Biology & Biochemistry 30, 393-402.
Swinnen, J., 1994. Evaluation of the use of a model rhizodeposition technique to separate root and
microbial respiration in soil. Plant and Soil 165, 89-101.
Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring soil microbial biomass C. Soil Biology & Biochemistry 19, 703-707.
Warembourg, F.R., Paul, E.A., 1977. Seasonal transfers of assimilated 14C in grassland: plant
production and turnover, soil and plant respiration. Soil Biology & Biochemistry 9, 295-301.
Werth, M., Kuzyakov, Y., 2006. Assimilate partitioning affects 13C fractionation of recently assimilated carbon in maize. Plant and Soil 284, 311-325.
Werth, M., Subbotina, I., Kuzyakov, Y., 2006. Three-source partitioning of CO2 efflux from soil
planted with maize by 13C natural abundance fails due to inactive microbial biomass. Soil
Biology & Biochemistry 38, 2772-2781.
Wu, J., Jörgensen, R.G., Pommerening, B., Chaussod, R., Brookes, P.C., 1990. Measurement of
soil microbial biomass-C by fumigation-extraction – an automated procedure. Soil Biology
& Biochemistry 22, 1167-1169.
Zibilske, L.M., 1994. Carbon Mineralization. In: Weaver, R.W., Angle, S., Bottomley, P., Bezdicek, D., Smith, S., Tabatabai, A., Wollum, A. (Eds.), Methods of Soil Analysis, Part 2,
Microbiological and Biochemical Properties. Soil Science Society of America Book Series,
No. 5, Soil Sci. Soc. Am., Inc., Madison, pp. 835-864.
6
Final conclusions and perspectives
108
6 Final conclusions and perspectives
Conclusions
Soils are an important source of CO2 and thus CO2 emissions from soils may contribute to the
global climatic change. In order to determine a soil’s function as a carbon source or sink, it is important to identify the sources of CO2 emissions. These sources can be directly plant-derived, so
that carbon is very shortly cycled between plants, soil, and atmosphere and no additional CO2 is
released into the atmosphere when compared to Net Primary Productivity. On the other hand,
rhizodeposits and plant residues get humified and become a part of soil organic matter (SOM). A
part of this carbon sink will be mineralised to CO2 and this fraction will increase by a change of
environmental conditions (e.g. increased temperature or high tillage agriculture), so that additional
SOM-derived CO2 will be released into the atmosphere. These main soil carbon pools and their related flows were considered in the present thesis in the rhizosphere of maize planted on C3 soil (Fig.
6-1).
Fig. 6-1: Carbon flows in the rhizosphere of maize planted on C3 soil. In every pool the carbon source is
indicated as C4 (maize-derived) or C3 (SOM-derived). Example δ13C values are shown from Chapter 3.
6 Final conclusions and perspectives
109
In the present thesis, partitioning the CO2 efflux from soil into root-derived contributions (including root and rhizomicrobial respiration) and SOM-derived contributions (i.e. SOM decomposition by soil microorganisms) was done by three different methods:
1) root exclusion, i.e. subtracting the CO2 emission from bare fallow soil from the one from
planted soil; this difference was defined as root-derived respiration (RDR), while the total
CO2 efflux from bare fallow soil was defined as SOM decomposition (SOMD),
14
2) C pulse labelling of maize shoots, chasing the tracer in soil compartments, and calculating
total root-derived CO2 by using an equation including total C and 14C in the maize shoots;
RDR was obtained by relating this result to the total CO2 efflux, SOMD was the difference
to 100%,
3) natural 13C labelling by planting maize on C3 soil, calculating RDR by a mass balance equation including δ13C values of CO2, maize roots, and SOM; SOMD was the difference to
100%.
The natural 13C labelling technique was compared to the root exclusion method under field conditions and to the 14C pulse labelling approach under controlled laboratory conditions (Table 6-1). In
both studies, a good correspondence was found between the root-derived CO2 contribution to soil
CO2 efflux of the root exclusion method (Chapter 5) and of the 14C pulse labelling approach (Chapter 4) when compared to the natural 13C labelling technique with fractionation. The 14C labelling
Table 6-1: Partitioning of CO2 efflux from soil planted with maize into four components. In the natural 13C
labelling approach 13C fractionations are included according to Table 6-2, i.e. mean fractionations in the
laboratory experiment (upper part) and seasonal fractionations in the field experiment (lower part).
Age
SOMD
RMR
RR
RDR
Method Chapter
%
July
July
September
September
14
16
16
22
22
28
28
34
34
40
40
12
24
31
17
6
30
23
12
-16
9
±
±
±
±
±
±
±
±
±
±
15
8
13
12
7
4
13
7
16
11
n.d.
-2 ± 2
n.d.
-3 ± 2
n.d.
4 ± 3
n.d.
2 ± 2
n.d.
5 ± 5
n.d.
78 ± 9
n.d.
86 ± 10
n.d.
66 ± 6
n.d.
86 ± 8
n.d.
86 ± 16
88
76
69
83
94
70
77
88
116
91
±
±
±
±
±
±
±
±
±
±
15
8
13
12
7
4
13
7
16
11
p C
13
n C
14
p C
13
n C
14
p C
13
n C
14
p C
13
n C
14
p C
13
n C
4
3+4
4
3+4
4
3+4
4
3+4
4
3+4
47
47
124
124
71
65
78
78
±
±
±
±
7
9
13
9
n.d.
4 ± 7
n.d.
8 ± 3
n.d.
31 ± 9
n.d.
14 ± 12
29
35
22
22
±
±
±
±
7
9
13
9
re
13
n C
re
13
n C
5
5
5
5
CO 2 pools: SOMD – CO 2 from soil organic matter decomposition, RMR – rhizomicrobial respiration,
RR – root respiration, RDR – root-derived respiration
methods: p 14C – 14C pulse labelling, n13C – natural 13C labelling, re – root exclusion
n.d. – not determined
110
6 Final conclusions and perspectives
approach, however, should only be used if the C amount in the shoot increases linearly, which was
true on the first four of five sampling days. Root-derived respiration was much lower in the field
study (22 to 35% of total CO2 efflux, Table 6-1 lower part) in comparison with the laboratory study
(69 to 94% of total CO2 efflux, Table 6-1 upper part). This difference was found because in the
laboratory CO2 was more precisely sampled from the complete rhizosphere soil, while in the field
the samples contained also large amounts of non-rhizosphere CO2. Higher contributions of rootderived CO2 in the laboratory in contrast to the field were also found in several other studies (see
Table 5-1) and this difference can be furthermore explained by the different environmental influences (soil moisture, temperature, etc.) between experiments under controlled and field conditions.
A further split-up of root-derived CO2 into root respiration (RR) and rhizomicrobial respiration
(RMR) is exceptionally difficult. In former studies (reviewed by Kuzyakov, 2006) this has mostly
been done by 14C labelling in a variety of approaches. The basic aim of the present thesis was to
verify the natural 13C labelling approach by Kuzyakov (2004; 2005). The δ13C values from the following four pools were required to partition the soil CO2 efflux: (1) total CO2 efflux (trapped as
carbonate in alkali), (2) maize roots, (3) soil organic matter, and (4) microbial biomass (determined
by chloroform fumigation-extraction). Results of former studies using 14C labelling techniques revealed that root-derived CO2 consists of 50% each for RR and RMR (see Table 5-1). This partitioning could not be confirmed by the natural 13C labelling used in this thesis (Table 6-1). Total CO2
efflux contributions of rhizomicrobial respiration (between 2 and 5% in the laboratory and between
4 and 8% in the field) were very low when compared with those of root respiration (between 66 and
86% in the laboratory and between 14 and 31% in the field). Although former studies are based on a
variety of assumptions, all of those studies determined much higher contributions of RMR than presented in this thesis. Consequently, a further task of the present study was to identify the main reasons for the inconsistent RMR contributions between our and former studies. As highlighted in
Chapters 3 and 5, one major problem of the natural 13C labelling approach by Kuzyakov (2004;
2005) is the small size of active microbial biomass. In this approach, microbial biomass extracts are
used as a substitute for microbially derived CO2. This substitution, however, was the main reason
for the underestimation of rhizomicrobial respiration in the present thesis. The discrepancy in this
substitution was that the δ13C values of the microbial biomass extract and of the microbially derived
CO2 did presumably not correspond. While the δ13C value of microbial biomass extract was dominated by the large amount of inactive or SOM-feeding microorganisms (which was proven by the
equality of microbial biomass δ13C values from maize and control treatments), the δ13C value of the
CO2 derived mainly from the smaller fraction of respiring active microorganisms, which fed mainly
on C4-derived rhizodeposits. In Chapters 3 and 5, we determined the C4 source consuming portion
of the active microbial biomass to be between 5 and 11% of the total microbial biomass. The total
active microbial biomass, however, would also consume a smaller proportion of C3-derived older
SOM (Fig. 6-1), hence, the total active microbial biomass would be slightly bigger and the passive
microbial biomass would be correspondingly smaller than estimated in this thesis.
6 Final conclusions and perspectives
111
The study with maize grown in nutrient solution (Chapter 2) and the control treatments with C3
soil without any plants (Chapters 3 to 5) revealed that 13C fractionation processes have to be considered when calculating below-ground CO2 fluxes and that this fractionation depends on partitioning of recent assimilates and on the nutrient status (Table 6-2). In a sensitivity analysis we showed
that fractionation has a big influence on RR and RMR partitioning results and that exact determination of 13C fractionation is essential for further calculations (Chapter 3). Several 13C fractionations
can be distinguished within the plant–soil carbon cycle. Following the 13C discrimination during
CO2 fixation by PEP carboxylase in maize (about 6‰ compared to δ13C of air CO2), further fractionation processes may occur during assimilate partitioning within the plant (O'Leary, 1981; Hobbie and Werner, 2004; Badeck et al., 2005), rhizodeposition (Yoneyama et al., 1997; Hobbie et al.,
2004; Cernusak et al., 2005), uptake by microorganisms (Ryan et al., 1995; Šantrůčková et al.,
2000; Potthoff et al., 2003), plant respiration (Lin and Ehleringer, 1997; Ghashghaie et al., 2003;
Klumpp et al., 2005), and SOM decomposition (Mary et al., 1992; Schweizer et al., 1999; Šantrůčková et al., 2000). These fractionations can be explained by three reasons: (1) different kinetics
of reactions with C isotopes, i.e. 12C reacting generally faster in chemical processes than 13C leading
to 13C-depleted products, (2) the difference in δ13C values of structural (lignin, lipids, etc.) and
transfer compounds (e.g. sucrose) leading to 13C-enriched roots compared to the shoots, and (3)
preferred utilization of 13C-enriched compounds (sugars, cellulose, hemicelluloses, etc.) by soil microorganisms resulting in their 13C enrichment.
The 13C enrichment of roots compared to shoots was observed in every study of this thesis but
was most pronounced for the mature plants in the field study (Table 6-2). Due to preferred utilizaTable 6-2: Fractionation of 13C between soil organic matter, microbial biomass, microbially derived CO2,
maize shoots and roots, root-derived CO2, and rhizodeposits.
Age
SOM-MB
MB-CO2
SOM-CO2
shoot-root
root-CO2
root-rhizod. Chapter
‰
NS
DNS
DW
July
September
19
24
29
19
24
29
19
24
29
n.d.
16
22
28
34
40
mean
n.d.
-4.0
-4.1
-2.0
-2.7
-3.2 ± 1.0
n.d.
-1.1
-2.0
-3.3
-1.6
-2.0 ± 0.9
-8.2
-5.1
-6.1
-5.3
-4.2
-5.2 ± 0.8
-0.2
-0.1
-0.2
+0.1
0
0
47
124
-1.2 ± 0.6
-1.7 ± 0.8
-0.6 ± 1.1
-3.7 ± 1.0
-1.4 ± 0.7
-5.5 ± 0.4
n.d.
-0.9 ± 0.1
n.d.
n.d.
n.d.
n.d.
-0.3 ± 0.1
n.d.
n.d.
-0.3 ± 0.1
n.d.
n.d.
-0.6 ± 0.2
±
±
±
±
±
±
0.2
0.4
0.1
0.1
0.1
0.1
+0.8
+1.0
+0.3
0
-0.5
-0.2
-0.7
-0.3
+0.2
±
±
±
±
±
±
±
±
±
0.5
0.4
0.2
0.7
0.1
0.5
0.3
0.3
0.5
-7.2
-5.2
-3.5
+2.2
+1.0
+2.4
+1.7
+1.7
+3.0
±
±
±
±
±
±
±
±
±
0.9
2.1
1.1
0.3
0.3
0.7
0.6
0.3
0.9
2
2
2
3+4
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
SOM – soil organic matter; MB – microbial biomass; NS – nutrient solution; DNS – 10 x diluted nutrient solution; DW – deionised water;
n.d. – not determined; mean – mean fractionation of days 22 to 40
5
112
6 Final conclusions and perspectives
tion of 13C-enriched compounds δ13C values of microbial biomass and of microbially derived CO2
increased during decomposition of SOM. If the reason for this δ13C increase in the process of microbial SOM decomposition were reaction kinetics, the resulting CO2 would have been 13C-depleted
compared to the digested substrate incorporated into the microbial biomass. This pattern, however,
was not found in any of our studies. Significantly 13C-depleted CO2 – probably due to kinetics –
was only found in the root respiration of the nutrient-rich treatment in Chapter 2. Since this depletion was only 0.7‰ on average, we did not consider it in the following chapters in line with most
rhizosphere CO2 studies (Cerling et al., 1991; Cheng, 1996; Lin and Ehleringer, 1997; Amundson et
al., 1998; Ekblad and Högberg, 2000; Fu and Cheng, 2002). Consequently, we only considered 13C
fractionations during microbial respiration, but not during root respiration.
In conclusion, the tested approach on CO2 efflux partitioning into three components (Kuzyakov,
2004; 2005) did not reflect the partitioning results of earlier studies due to the low sensitivity of the
method and the small amount of active microbial biomass consuming C4-derived rhizodeposits, but
dominating microbial CO2. The application of the 14C pulse labelling, natural 13C labelling, and root
exclusion approaches, however, is suitable to partition the CO2 efflux into root- and SOM-derived
contributions. The two isotopic methods may also be used to determine these contributions in the
microbial biomass. New approaches on CO2 efflux partitioning into three sources have to be developed excluding the usage of microbial biomass extracts.
Perspectives
The natural 13C labelling, artificial 14C pulse labelling, and root exclusion methods have been used
to to separate root- and SOM-derived C contributions to the total CO2 efflux or to the soil microbial
biomass. In addition, molecular methods like biomarkers could be used to identify C3- and C4derived C contributions to certain community parts of the soil microbial biomass (δ13C of phospholipid fatty acids (PLFA)), to determine plant-derived or microbial residues (individual sugars), or to
determine bacterial or fungal residues (individual amino sugars) in the soil (reviewed by Glaser,
2005). Rhizodeposit-feeding active microorganisms could be determined by a PLFA analysis, for
instance, and an average δ13C value of these organisms could be used in the approach by Kuzyakov
(2004; 2005) on partitioning the soil CO2 efflux. Inactive microorganisms would then be excluded
from calculations. Thus, further research has to be done on this CO2 efflux partitioning method and
on identifying root- and SOM-derived C contributions by combination of molecular and isotopic
methods. It still remains problematic, however, to use the 13C natural abundance in this research
area due to its low sensitivity when compared with artificial 14C or 13C labelling.
References
Amundson, R., Stern, L., Baisden, T., Wang, Y., 1998. The isotopic composition of soil and soilrespired CO2. Geoderma 82, 83-114.
6 Final conclusions and perspectives
113
Badeck, F.-W., Tcherkez, G., Nogués, S., Piel, C., Ghashghaie, J., 2005. Post-photosynthetic
fractionation of stable carbon isotopes between plant organs – a widespread phenomenon.
Rapid Communications in Mass Spectrometry 19, 1381-1391.
Cerling, T.E., Solomon, D.K., Quade, J., Bowman, J.R., 1991. On the isotopic composition of
carbon in soil carbon dioxide. Geochimica et Cosmochimica Acta 55, 3404-3405.
Cernusak, L.A., Farquhar, G.D., Pate, J.S., 2005. Environmental and physiological controls over
oxygen and carbon isotope composition of Tasmanian blue gum, Eucalyptus globulus. Tree
Physiology 25, 129-146.
Cheng, W., 1996. Measurement of rhizosphere respiration and organic matter decomposition using
natural 13C. Plant and Soil 183, 263-268.
Ekblad, A., Högberg, P., 2000. Analysis of δ13C of CO2 distinguishes between microbial respiration of added C4-sucrose and other soil respiration in a C3-ecosystem. Plant and Soil 219,
197-209.
Fu, S., Cheng, W., 2002. Rhizosphere priming effects on the decomposition of soil organic matter
in C4 and C3 grassland soils. Plant and Soil 238, 289-294.
Ghashghaie, J., Badeck, F.-W., Lanigan, G., Nogués, S., Tcherkez, G., Deléens, E., Cornic, G.,
Griffiths, H., 2003. Carbon isotope fractionation during dark respiration and photorespiration in C3 plants. Phytochemistry Reviews 2, 145-161.
Glaser, B., 2005. Compound-specific stable-isotope 13C analysis in soil science. Journal of Plant
Nutrition and Soil Science 168, 633-648.
Hobbie, E.A., Werner, R.A., 2004. Intramolecular, compound-specific, and bulk carbon isotope
patterns in C3 and C4 plants: a review and synthesis. New Phytologist 161, 371-385.
Hobbie, E.A., Johnson, M.G., Rygiewicz, P.T., Tingey, D.T., Olszyk, D.M., 2004. Isotopic estimates of new carbon inputs into litter and soils in a four-year climate change experiment
with Douglas-fir. Plant and Soil 259, 331-343.
Klumpp, K., Schäufele, R., Lötscher, M., A., L.F., Feneis, W., Schnyder, H., 2005. C-isotope
composition of CO2 respired by shoots and roots: fractionation during dark respiration?
Plant, Cell and Environment 28, 241-250.
Kuzyakov, Y., 2004. Separation of root and rhizomicrobial respiration by natural 13C abundance:
theoretical approach, advantages, and difficulties. Eurasian Soil Science 37, S79-S84.
Kuzyakov, Y., 2005. Theoretical background for partitioning of root and rhizomicrobial respiration
by δ13C of microbial biomass. European Journal of Soil Biology 41, 1-9.
Kuzyakov, Y., 2006. Sources of CO2 efflux from soil and review of partitioning methods. Soil Biology & Biochemistry 38, 425-448.
Lin, G., Ehleringer, J.R., 1997. Carbon isotope fractionation does not occur during dark respiration in C3 and C4 plants. Plant Physiology 114, 391-394.
Mary, B., Mariotti, A., Morel, J.L., 1992. Use of 13C variations at natural abundance for studying
the biodegradation of root mucilage, roots and glucose in soil. Soil Biology & Biochemistry
24, 1065-1072.
O'Leary, M.H., 1981. Carbon isotope fractionation in plants. Phytochemistry 20, 553-567.
Potthoff, M., Loftfield, N., Buegger, F., Wick, B., John, B., Jörgensen, R.G., Flessa, H., 2003.
The determination of δ13C in soil microbial biomass using fumigation-extraction. Soil Biology & Biochemistry 35, 947-954.
Ryan, M.C., Aravena, R., Gillham, R.W., 1995. The use of 13C natural abundance to investigate
the turnover of the microbial biomass and active fractions of soil organic matter under two
114
6 Final conclusions and perspectives
tillage treatments. In: Lal, R., Kimble, J., Levine, E., Stewart, B.A. (Eds.), Soils and Global
Change, CRC Press, Boca Raton, pp. 351-360.
Šantrůčková, H., Bird, M.I., Lloyd, J., 2000. Microbial processes and carbon-isotope fractionation in tropical and temperate grassland soils. Functional Ecology 14, 108-114.
Schweizer, M., Fear, J., Cadisch, G., 1999. Isotopic (13C) fractionation during plant residue decomposition and its implications for soil organic matter studies. Rapid Communications in
Mass Spectrometry 13, 1284-1290.
Yoneyama, T., Handley, L.L., Scrimgeour, C.M., Fisher, D.B., Raven, J.A., 1997. Variations of
the natural abundances of nitrogen and carbon isotopes in Triticum aestivum, with special
reference to phloem and xylem exudates. New Phytologist 137, 205-213.
115
7
Summary
Soils contain the largest terrestrial carbon (C) pool and thus play an important role in the global
carbon cycle. Root respiration, microbial decomposition of plant residues and rhizodeposits, and
mineralisation of soil organic matter (SOM) lead to the release of CO2 into the atmosphere. This
atmospheric C in turn is assimilated by plants, returned into the soil by rhizodeposition and litter
input, and incorporated into SOM by biological, chemical and physical processes. On the issue of
increasing atmospheric CO2 concentrations much interest exists in finding out whether and under
which conditions soils act as net sink or source in the carbon cycle. Separation of the CO2 efflux
from soil into root-derived and microbially derived components and further partitioning of the latter
into microbial respiration derived from rhizodeposits and from SOM is of major importance in answering this question.
In the present thesis these fluxes were examined by planting maize (Zea mays L.), a C4 plant, on
C3 soil either under controlled laboratory conditions or in the field. Total C translocated below
ground by plants during the growing season or during an experimental period was estimated by this
natural 13C labelling. In addition, recently assimilated carbon was chased in several pools by artificial 14C pulse labelling of the shoots. This thesis is based on three experiments.
The dependence of 13C fractionation on maize assimilate partitioning between shoot, root,
rhizodeposits, and root-respired CO2 was investigated by coupling 13C natural abundance and 14C
pulse labelling under three nutrient treatments. Increasing amounts of recently assimilated C in the
roots from 8% to 13% of recovered 14C (nutrient solution < 10 x diluted nutrient solution < deionised water) led to a 0.6‰ 13C enrichment from high to low nutrient treatments. This increase in the
δ13C value was reflected in the root-derived CO2, where the 13C enrichment between high and low
nutrient treatments was at 1.6‰. δ13C of CO2 evolved by root respiration was generally similar to
that of the roots, i.e. the respired CO2 was not exceeding a 0.7‰ 13C depletion compared to roots.
Consequently, 13C discrimination between plant and soil pools is not constant but strongly depends
on partitioning of recently assimilated C between these pools.
116
7 Summary
After the determination of influences on 13C fractionation between plant parts and below-ground
carbon pools, the approach of Kuzyakov (2004 1 , 2005 2 ) was examined, which uses the natural 13C
labelling technique to partition the CO2 efflux from soil into three fluxes: (1) root respiration (RR),
(2) rhizomicrobial respiration (RMR), and (3) microbial SOM decomposition (SOMD). Without
considering any 13C fractionation, RR, RMR, and SOMD amounted 91%, 4%, and 5%, respectively. This approach strongly overestimated RR and underestimated RMR and SOMD when compared with former studies mostly based on 14C labelling methods. The consideration of a 2.0‰ 13C
fractionation between microbial biomass (MB) and CO2 changed the proportions of RR and RMR
by only 4% and did not affect SOMD. An active portion of 37% to total MB in the rhizosphere was
calculated, assuming the δ13C of active MB to be equal to the δ13C of maize roots and the δ13C of
passive MB to be equal to the δ13C of SOM (including fractionation between MB and CO2). In conclusion, the method for partitioning total below-ground CO2 efflux into three sources using a natural
13
C labelling technique did not produce comparable RR and RMR contributions to former studies
due to the small proportion of active MB in the rhizosphere. This small active fraction led to a discrepancy between δ13C values of microbial biomass and of microbially respired CO2.
In the same experiment, we compared the amounts of root-derived carbon estimated by two isotopic approaches: (1) natural 13C labelling and (2) artificial 14C pulse labelling. The root-derived
carbon in the total CO2 efflux was between 69% and 94% in the 14C labelling approach and between
86% and 94% in the natural 13C labelling approach. At an assumed 13C fractionation between SOM
and CO2 of 5.2‰, the root-derived contribution to CO2 ranged from 70% to 88% and was much
closer to the results of the 14C labelling approach. Root-derived contributions to the MB carbon
ranged from 2% to 9% using 14C labelling and from 16% to 36% using natural 13C labelling. At a
3.2‰ 13C fractionation between SOM and MB, both labelling approaches yielded an equal contribution of root-derived C to the microbial biomass. Both approaches may therefore be used to partition CO2 efflux and to quantify the MB carbon sources. However, the assumed 13C fractionation
strongly affected the contributions of individual C sources. In contrast to the underestimation of
SOMD when compared with literature results in the former paragraph, the natural 13C labelling
technique has been confirmed in partitioning SOM- and root-derived CO2 by the 14C pulse labelling
method. The contribution of RMR, however, still remained remarkably low when compared to former studies and this questions the suitability of the natural 13C labelling technique to determine this
particular CO2 efflux contribution.
In a field experiment, root-derived CO2 contributions were estimated by the root exclusion
(comparison of planted and bare fallow soil) and by the natural 13C labelling technique tested earlier
in the laboratory. In both approaches, CO2 from SOM decomposition dominated 65% to 78% of the
total CO2 efflux during the growing season. Without considering 13C fractionation, rhizomicrobial
1
2
Kuzyakov, Y., 2004. Separation of root and rhizomicrobial respiration by natural 13C abundance: theoretical approach, advantages, and difficulties. Eurasian Soil Science 37, S79-S84.
Kuzyakov, Y., 2005. Theoretical background for partitioning of root and rhizomicrobial respiration by δ13C of microbial biomass. European Journal of Soil Biology 41, 1-9.
7 Summary
117
respiration was very low compared to other studies. At the end of the growing season, when considering high isotopic fractionations between SOM, MB, and CO2, however, root and rhizomicrobial
respiration amounted to 64% and 36% of root-derived CO2, respectively. This relationship was
closer to the 50:50% partitioning described in the literature than without considering fractionation.
Consequently, 13C fractionation processes must be taken into account when calculating CO2 efflux
partitioning from soil. Both methods – root exclusion and natural 13C labelling – showed the same
CO2 partitioning results when 13C isotopic fractionation during microbial respiration was considered.
In conclusion, the tested approach on CO2 efflux partitioning did not reflect the partitioning results of RR and RMR in earlier studies. This difference was caused by the low sensitivity of natural
13
C labelling and the small amount of active microbial biomass consuming C4-derived rhizodeposits, but dominating microbial CO2. Root- and SOM-derived C in the CO2, however, was successfully determined in the laboratory (by artificial 14C and natural 13C labelling) and in the field (by
root exclusion and natural 13C labelling). The consideration of 13C fractionation was of major importance in all our studies. New methods on CO2 efflux partitioning – especially on RR and RMR
separation – have to be developed based on a combination of classical and isotopic techniques.
118
8
Zusammenfassung
Böden beinhalten den größten terrestrischen Kohlenstoffpool und spielen daher eine wichtige Rolle
im globalen Kohlenstoffkreislauf. Wurzelatmung, mikrobieller Abbau von Pflanzenresten und Rhizodepositen und Mineralisierung der organischen Bodensubstanz (SOM) führen zur Freisetzung
von CO2 in die Atmosphäre. Dieser atmosphärische Kohlenstoff (C) wiederum wird von den Pflanzen assimiliert, durch Rhizodeposition und Streueintrag dem Boden wieder zugeführt und durch
biologische, chemische und physikalische Prozesse in die SOM eingebaut. Vor dem Hintergrund
ansteigender atmosphärischer CO2-Konzentrationen besteht großes Interesse daran, herauszufinden,
ob und unter welchen Bedingungen Böden als Nettoquelle oder -senke im Kohlenstoffkreislauf agieren. Die Unterteilung des CO2-Effluxes aus dem Boden in wurzel- und mikroorganismenbürtige
Anteile und eine weitere Aufteilung des letzteren in die mikrobielle Veratmung der Rhizodeposite
und der SOM ist von größter Bedeutung bei der Beantwortung dieser Fragestellung.
Diese Flüsse wurden in der vorliegenden Doktorarbeit durch die Anpflanzung von Mais (Zea
mays L.), einer C4-Pflanze, auf einem C3-Boden unter kontrollierten Laborbedingungen oder im
Feldversuch untersucht. Der durch die Pflanzen in den Boden verlagerte Gesamtkohlenstoff während der Vegetationsperiode oder während einer Versuchsdauer wurde durch natürliche 13CMarkierung abgeschätzt. Zusätzlich wurde der kürzlich assimilierte Kohlenstoff durch künstliche
14
C-Pulsmarkierung der Sprosse in verschiedenen Pools verfolgt. Die vorliegende Dissertation basiert auf drei Experimenten.
Die Abhängigkeit der 13C-Fraktionierung von der Maisassimilatverteilung zwischen Spross,
Wurzel, Rhizodepositen und wurzelbürtigem CO2 wurde durch Kombination von natürlicher 13CAbundanz und 14C-Pulsmarkierung unter drei Nährstoffvarianten untersucht. Zunehmende Mengen
von kürzlich assimiliertem C in den Wurzeln von 8 % auf 13 % des wiedergefunden 14C (Nährlösung < 10-fach verdünnte Nährlösung < deionisiertes Wasser) führten zu einer 13C-Anreicherung
von 0,6 ‰ von der hohen zur niedrigen Nährstoffvariante. Diese Zunahme des δ13C-Wertes wurde
auch im wurzelbürtigen CO2 gefunden, wo die 13C-Anreicherung zwischen hohen und niedrigen
Nährstoffvarianten bei 1,6 ‰ lag. Der δ13C-Wert des CO2 aus der Wurzelatmung stimmte mit dem
der Wurzeln im allgemeinen überein, d. h. das veratmete CO2 war gegenüber den Wurzeln maximal
um 0,7 ‰ 13C-abgereichert. Folglich ist die 13C-Diskriminierung zwischen Pflanzen- und Bodenpools nicht konstant sondern stark von der Verteilung des kürzlich assimilierten C zwischen diesen
Pools abhängig.
8 Zusammenfassung
119
Nach der Bestimmung der Einflüsse auf die 13C-Fraktionierung zwischen Pflanzenteilen und unterirdischen Kohlenstoffpools wurde der Ansatz von Kuzyakov (2004 1 , 2005 2 ) untersucht, der die
natürliche 13C-Markierungstechnik zur Unterteilung des CO2-Effluxes aus dem Boden in folgende
drei Flüsse verwendet: (1) Wurzelatmung (RR), (2) rhizomikrobielle Atmung (RMR) und (3)
mikrobieller Abbau der SOM (SOMD). Ohne Berücksichtigung der 13C-Fraktionierung betrugen
RR, RMR und SOMD jeweils 91 %, 4 % bzw. 5 %. Diese Methode überschätzte die RR und unterschätzte RMR und SOMD deutlich im Vergleich zu früheren Studien, die meistens auf 14CMarkierungsmethoden basierten. Der Einbezug einer 13C-Fraktionierung von 2,0 ‰ zwischen
mikrobieller Biomasse (MB) und CO2 veränderte die Anteile von RR und RMR nur um 4 % und
beeinflusste SOMD nicht. Unter der Annahme, dass der δ13C-Wert der aktiven MB gleich dem der
Maiswurzeln und der δ13C-Wert der passiven MB gleich dem der SOM (mit Fraktionierung zwischen MB und CO2) sein soll, wurde ein aktiver Anteil von 37 % an der gesamten MB in der Rhizosphäre berechnet. Die Methode zur Aufteilung des unterirdischen CO2-Gesamteffluxes in drei
Quellen mit Hilfe der natürlichen 13C-Markierungstechnik lieferte aufgrund des kleinen Anteils
aktiver MB in der Rhizosphäre schließlich keine RR- und RMR-Beiträge, die zu früheren Studien
vergleichbar sind. Diese kleine aktive Fraktion führte zu einer Diskrepanz zwischen den δ13CWerten der mikrobiellen Biomasse und des mikrobiell veratmeten CO2.
In demselben Experiment haben wir die wurzelbürtigen Kohlenstoffmengen miteinander verglichen, die durch zwei isotopische Ansätze berechnet wurden: (1) natürliche 13C-Markierung und (2)
künstliche 14C-Pulsmarkierung. Der wurzelbürtige Kohlenstoff im CO2-Gesamtefflux lag zwischen
69 % und 94 % im 14C-Markierungsansatz und zwischen 86 % und 94 % im natürlichen 13CMarkierungsansatz. Unter Annahme einer 13C-Fraktionierung zwischen SOM und CO2 von 5,2 ‰
lag der wurzelbürtige Beitrag zum CO2 zwischen 70 % und 88% und war viel näher an den Ergebnissen des 14C-Markierungsansatzes. Wurzelbürtige Beiträge zum Kohlenstoff der MB lagen zwischen 2 % und 9 % berechnet durch 14C-Markierung und zwischen 16 % und 36 % berechnet durch
natürliche 13C-Markierung. Bei einer 13C-Fraktionierung zwischen SOM und MB von 3,2 ‰ erzielten beide Markierungsmethoden gleiche Beiträge des wurzelbürtigen C zur mikrobiellen Biomasse.
Beide Ansätze können daher zur Aufteilung des CO2-Effluxes und zur Quantifizierung von Kohlenstoffquellen der MB verwendet werden. Die angenommene 13C-Fraktionierung beeinflusste die Beiträge der einzelnen C-Quellen allerdings deutlich. Im Gegensatz zur Unterschätzung von SOMD im
Vergleich zu den Ergebnissen aus der Literatur aus dem vorherigen Absatz, wurde die natürliche
13
C-Markierungsmethode zur Aufteilung von SOM- und wurzelbürtigem CO2 durch die 14CPulsmakierungsmethode bestätigt. Im Vergleich zu früheren Studien blieb der Beitrag der RMR
1
2
Kuzyakov, Y., 2004. Separation of root and rhizomicrobial respiration by natural 13C abundance: theoretical approach, advantages, and difficulties. Eurasian Soil Science 37, S79-S84.
Kuzyakov, Y., 2005. Theoretical background for partitioning of root and rhizomicrobial respiration by δ13C of microbial biomass. European Journal of Soil Biology 41, 1-9.
120
8 Zusammenfassung
allerdings immer noch sehr klein, was die Verwendbarkeit der natürlichen 13C-Markierungstechnik
zur Bestimmung dieses speziellen CO2-Effluxbeitrags in Frage stellt.
In einem Feldexperiment wurden die Beiträge von wurzelbürtigem CO2 mit der Wurzelausschlussmethode (Vergleich von bepflanztem Boden und Schwarzbracheboden) und mit der natürlichen 13C-Markierungsmethode, die schon zuvor im Labor überprüft wurde, abgeschätzt. Während
der Vegetationsperiode dominierte das CO2 aus dem Abbau der SOM in beiden Ansätzen 65 % bis
78 % des CO2-Gesamteffluxes. Ohne Betrachtung von 13C-Fraktionierung war die rhizomikrobielle
Atmung im Vergleich zu anderen Studien sehr niedrig. Unter Einbezug hoher isotopischer Fraktionierungen zwischen SOM, MB und CO2 am Ende der Vegetationsperiode betrugen die Wurzel- und
rhizomikrobielle Atmung allerdings 64 % bzw. 36 %. Dieses Verhältnis war näher an der 50:50 %Aufteilung aus der Literatur als ohne Einbezug der Fraktionierung. 13C-Fraktionierungsprozesse
müssen folglich bei der Berechnung der Aufteilung des CO2-Effluxes aus dem Boden berücksichtigt werden. Beide Methoden, d. h. Wurzelausschluss- und natürliche 13C-Markierungsmethode,
lieferten gleiche CO2-Aufteilungsergebnisse, wenn die isotopische 13C-Fraktionierung während der
mikrobiellen Atmung berücksichtigt wurde.
Der überprüfte Ansatz zur Aufteilung des CO2-Effluxes gab schließlich nicht die Unterteilungen
in RR und RMR aus früheren Studien wieder. Diese Unstimmigkeit ist auf die geringe Sensitivität
der natürlichen 13C-Markierung und die geringen Anteile an aktiver mikrobieller Biomasse, die C4bürtige Rhizodeposite konsumiert, aber gleichzeitig das mikrobielle CO2 dominiert, zurückzuführen. Wurzel- und SOM-bürtiger C im CO2 konnte jedoch erfolgreich im Labor (durch künstliche
14
C- und natürliche 13C-Markierung) und Feld (durch Wurzelausschluss und natürliche 13CMarkierung) bestimmt werden. Der Einbezug der 13C-Fraktionierung war von größter Bedeutung in
all unseren Studien. Neue Methoden zur Aufteilung des CO2-Effluxes, besonders zur Trennung von
RR und RMR, müssen durch Kombination von klassischen und isotopischen Methoden entwickelt
werden.
121
Acknowledgements
First, I would like to thank Prof. Dr. Yakov Kuzyakov as supervisor of this thesis for his support
and constructive criticism. I am particularly grateful for the scientific freedom that he gave me, his
confidence in my competence and the valuable advice in writing research papers.
I am also grateful to Prof. Dr. Manfred Küppers for accepting to act as co-reviewer. I highly appreciate Prof. Dr. Karl Stahr, who accomodated me at the Institute of Soil Science and Land Evaluation
after the group of my supervisor had moved to the University of Bayreuth.
Many thanks to Elke Feiertag, Gabriela Bermejo Dominguez, Irina Subbotina, Melanie Wagner,
and Coral Monje Delgadillo for their help with the laboratory and field work.
I thank Dr. Vadim Cercasov, formerly head of the Central Isotope Laboratory at the University of
Hohenheim, for his friendly assistance in working with 14C.
Thanks also to Dr. Wolfgang Armbruster and Elke Dachtler for their kind support in providing time
for 13C measurements at the isotope ratio mass spectrometer.
I thank Herbert Stelz for his useful tips and his help in the field plot maintenance.
Thanks to Dr. Sven Marhan for the usage of the equipment of the Soil Biology section.
I am grateful to Dr. Michael Stachowitsch and Leonidas Kleanthous for critically reading and improving the text.
Special thanks to my colleagues and friends Holger Fischer and Katja Schneckenberger who always
had time for discussions and corrections of my work.
I am most grateful to my parents and grandparents for their interest in my research and their support
in every way. My major thank, however, is addressed to my girlfriend Annika, who always gave me
motivation to get the work finished.
This work was financially supported by the German Research Foundation (DFG).
122
Curriculum vitae
Martin Werth
Date of birth:
Place of birth:
Marital status:
09 February 1976
Haan/Rhineland, Germany
single
Vocational experience
July 2006 to
December 2006
January 2004 to
June 2006
June 2003 to
December 2003
January 2001 to
September 2002
September 1999
Research fellow at the Department of Agro-ecosystem Research,
Prof. Dr. Yakov Kuzyakov, University of Bayreuth, Germany, place of work:
Stuttgart-Hohenheim, Germany
Research fellow in the group of PD Dr. Yakov Kuzyakov,
Institute of Soil Science and Land Evaluation, University of Hohenheim,
Stuttgart, Germany
Research assistant in the group of Prof. Dr. Gabriele Broll
(Department of Geo- and Agro-ecology), Institut für Strukturforschung und
Planung in agrarischen Intensivgebieten (ISPA), University of Vechta, Germany
Research assistant in the group of Prof. Dr. Werner Kuhn,
Institute for Geoinformatics, University of Münster, Germany
Research assistant in the GIS-department, Wupperverband, Wuppertal, Germany
Higher education
Since April 2004
28 April 2003
October 2002 to
March 2003
October 1996 to
April 2003
PhD student in biology at the University of Hohenheim, Stuttgart, Germany,
supervisor: Prof. Dr. Yakov Kuzyakov
Graduation as ‘Landscape-ecologist’ (Diplom, i.e. Master equivalent, grade:
1.3)
Thesis on the topic ‘Modelling of soil organic matter and nutrients of
grassland sites in South-western Germany’, supervisor: Prof. Dr. Gabriele
Broll
Studies of landscape ecology at the University of Münster, Germany
main subjects: soil science, geoinformatics
subsidiary subjects: plant sciences, chemistry
Abroad experience
July 2001 to
September 2001
October 1998 to
June 1999
Placement at the Institut National de la Recherche Agronomique
(INRA) in Versailles, France, Task: Analysis of heavy metals
Geography studies at the University of Durham, England,
ERASMUS-scholarship
Education and Civilian Service
1995 – 1996
1986 – 1995
1982 – 1986
Civilian service in a hospital, Wuppertal, Germany
Grammar school, Wuppertal, Germany (Grade: 2.1)
Primary school, Wuppertal, German