Short communication
Soil macroaggregate turnover and microaggregate formation:
a mechanism for C sequestration under no-tillage agriculture
J. Six*, E.T. Elliott1, K. Paustian
Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, CO 80523, USA
Received 27 March 2000; received in revised form 4 July 2000; accepted 29 July 2000
Abstract
Soil disturbance from tillage is a major cause of organic matter depletion and reduction in the number and stability of soil aggregates when
native ecosystems are converted to agriculture. No-till (NT) cropping systems usually exhibit increased aggregation and soil organic matter
relative to conventional tillage (CT). However, the extent of soil organic matter changes in response to NT management varies between soils
and the mechanisms of organic matter stabilization in NT systems are unclear. We evaluated a conceptual model which links the turnover of
aggregates to soil organic matter dynamics in NT and CT systems; we argue that the rate of macroaggregate formation and degradation (i.e.
aggregate turnover) is reduced under NT compared to CT and leads to a formation of stable microaggregates in which carbon is stabilized and
sequestered in the long term. Therefore, the link between macroaggregate turnover, microaggregate formation, and C stabilization within
microaggregates partly determines the observed soil organic matter increases under NT. q 2000 Elsevier Science Ltd. All rights reserved.
Keywords: Aggregate turnover; Carbon sequestration; No-tillage; Particulate organic matter
An accumulation of soil organic matter under no-tillage
(NT) compared to conventional tillage (CT) confers impor-
tant improvements in soil quality, soil fertility and seques-
tration of carbon. Mechanisms by which reduced tillage
decreases soil organic matter decomposition (e.g. reduced
soil disturbance, redistribution of residues) have been
known for some time (Oades, 1984; Paustian et al., 1997)
but the ultimate mechanisms have not been well elucidated.
It has been established that the inclusion of organic materi-
als within soil aggregates reduces their decomposition rate
(Oades, 1984; Elliott and Coleman, 1988). Increases in
aggregation concomitant with increases in organic C have
been observed in NT systems (Paustian et al., 2000; Six et
al., 2000). Tillage has been found to induce a loss of C-rich
macroaggregates and a gain of C-depleted microaggregates
(Six et al., 2000). However, this decrease in macroaggre-
gates cannot explain the total C loss associated with tillage.
Six et al. (1998, 1999a) suggested that increased macroag-
gregate turnover under CT is a primary mechanism causing
decreases of soil C. Differential macroaggregate turnover
induced by tillage and its influence on SOM dynamics are
represented in a conceptual model (Fig. 1) we developed to
explain differences in C-sequestration between NT and CT.
The model forms a basis for the hypotheses tested in this
study.
At time 1 (t1), macroaggregates (250–2000 mm) are
formed around fresh residue which then becomes coarse
intra-aggregate particulate organic matter (iPOM) (Fig. 1).
The fresh residue induces the formation of macroaggregates
because it is a C source for microbial activity and the
production of microbial-derived binding agents (Golchin
et al., 1994; Jastrow, 1996; Six et al., 1999a). The model
assumes similar rates of macroaggregate formation in NT
versus CT (at t1) because the residue input is often similar in
both management systems (Paul et al., 1997). Six et al.
(1999a) found that the proportion of crop-derived C (i.e.
newer C) relative to native grassland C (i.e. older C) was
similar in NT and CT macroaggregates, confirming that the
rate of macroaggregate formation was similar in both
management systems.
Fine iPOM within a macroaggregate is derived from the
decomposition and subsequent fragmentation of coarse
iPOM (t1 to t2; Fig. 1). This process is supported by the obser-
vations that decomposition causes decreases in POM size
(Guggenberger et al., 1994) and that fine iPOM is older than
coarse iPOM (Six et al., 1998). Consequently, fine iPOM
concentration is expected to increase with macroaggregate
Soil Biology & Biochemistry 32 (2000) 2099–2103
0038-0717/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0038-0717(00)00179-6
www.elsevier.com/locate/soilbio
* Corresponding author. Tel.: 11-970-491-5599; fax: 11-970-491-1965.
E-mail address: johan@nrel.colostate.edu (J. Six).
1 Present address: School of Natural Resource Sciences, University of
Nebraska at Lincoln, Lincoln, NE 68583-0758, USA.
age. Therefore, an abundance of older macroaggregates
suggests a slower macroaggregate turnover and the ratio of
fine iPOM to coarse iPOM within macroaggregates can be
used as a relative measure of the turnover of macroaggregates.
At four agricultural experiment sites this ratio was approxi-
mately twice
2:10 ^ 0:30 as large in NT as in CT (Fig. 2),
suggesting that macroaggregate turnover is approximately
twice as slow in NT compared to CT.
As fine iPOM is formed it gradually becomes encrusted
with clay particles and microbial products (t2 to t3) to form
microaggregates within macroaggregates (Six et al., 1998,
1999a). In isotope tracer studies, labeled C was redistributed
from macroaggregates to microaggregates with time (Angers
et al., 1997; Gale et al., 2000) suggesting that microaggregates
are formed within macroaggregates (Oades, 1984; Elliott and
Coleman, 1988). Eventually, the binding agents in macroag-
gregates degrade, resulting in loss of macroaggregate stability
(t4) and the release of stable microaggregates, which become
the building blocks for the next cycle of macroaggregate
formation (Tisdall and Oades, 1982).
The model suggests that the increases in macroaggregate
turnover induced by tillage yields fewer new free microag-
gregates in CT compared to NT at t4 (Fig. 1). Supporting
this, less accumulation of crop-derived C in free micro-
aggregates has been observed in CT compared to NT
(Six et al., 1999a). This incorporation of new C into free
microaggregates is an important factor contributing to C-
sequestration (Skjemstad et al., 1990) since C contained in
J. Six et al. / Soil Biology & Biochemistry 32 (2000) 2099–21032100
Fig. 1. This conceptual model shows the “life cycle” of a macroaggregate and the formation of microaggregates. These processes are the result of the
interrelationship between the turnover of macroaggregates, the turnover of SOM and controlling factors such as disturbance (i.e. tillage). Turnover occurs
through time as an aggregate is formed, becomes unstable and is eventually disrupted. Disturbances such as tillage shortcut this “life cycle” of macroaggregates
and diminish the formation rate of new microaggregates and thus the sequestration of C within microaggregates and the soil as a whole. (iPOM intra-
aggregate particulate organic matter; POM particulate organic matter; t time; process rate modifier).
free microaggregates has a slower turnover than C in macro-
aggregates (Jastrow et al., 1996).
Based on the accumulated understanding embodied in
our conceptual model, we advanced the following two
hypotheses.
H1: in NT soils, there are more microaggregates contained
within macroaggregates and there is more fine iPOM-C
within these microaggregates than in CT soils.
H2: in NT soils, the SOM-C and iPOM-C of microaggre-
gates that are contained within macroaggregates are
younger than in CT soils.
The rationale for these two hypotheses is as follows. A slow
macroaggregate turnover in NT allows time for the formation
of fine iPOM from recent crop-derived coarse iPOM and the
subsequent encapsulation of this fine iPOM by mineral parti-
cles and microbial byproducts to form stable microaggregates
containing young crop-derived C. In contrast, the turnover of
macroaggregates in CT is faster, providing less opportunity for
the formation of crop-derived fine iPOM and stable microag-
gregates. Thus the organic matter contained in microaggre-
gates of CT soils is dominated by older C derived from the
pre-cultivation vegetation.
Direct testing of these hypotheses was accomplished by
isolating microaggregates held within macroaggregates
using a new method. Soil samples (0–5 cm) were taken at
a long-term tillage experiment located at Sidney, NE. The
experiment is on a Duroc loam, fine silty, mixed, mesic
Pachic Haplustoll and includes NT and CT under wheat
fallow, and a native grassland treatment. Prior to establish-
ment of the experiment (in 1969) the entire area was native
grassland that had never been cultivated. To isolate micro-
aggregates held within macroaggregates, we built a device
to completely break up macroaggregates while minimizing
the break down of the released microaggregates. We only
analyzed macroaggregates from the 0–5 cm layer because
we did not observe any significant differences between NT
and CT in aggregate carbon and fine iPOM-C in the 5–
20 cm layer at this site (Six et al., 1999a). Macroaggregates
(10 g) were immersed in deionized water on top of a 250-
mm mesh screen and gently shaken with 50 glass beads
(dia. 4 mm). Continuous and steady water flow through
the device ensured that microaggregates were immediately
flushed onto a 53-mm sieve and were not exposed to any
further disruption by the beads. After all macroaggregates
were broken up, the material on the 53-mm sieve was sieved
to ensure that the isolated microaggregates were water-
stable. The inter-microaggregate POM retained together
with the microaggregates on the sieve was isolated by
density flotation in 1.85 g cm23 sodium polytungstate (Six
et al., 1998). After density flotation, the microaggregates
were dispersed in 5 g hexametaphosphate per liter and
intra-microaggregate POM isolated by sieving (Six et al.,
1998). To avoid cross contamination of C, and especially
13C, among samples, sodium polytungstate was recycled
according to Six et al. (1999b). The proportion of micro-
aggregate (53–250 mm) weight within macroaggregates
(250–2000 mm) was calculated as:
microaggregate weight 2 weight of 53–250 mm sized sand
macroaggregate weight 2 weight of 250–2000 mm sized sand
1
The weights of macro- and microaggregates were corrected
for the sand content of the same size as the aggregates
because sand of the same size as the aggregate is usually
not a part of an aggregate and should consequently not be
weighed as an aggregate. Carbon concentrations were
measured with a LECO CHN-1000 analyzer and carbon-
isotope ratios were determined using a Carlo Erba NA
1500 CN analyzer coupled to a Micromass VG isochrom-
EA masspectrometer.
Our hypotheses were corroborated with data from this new
approach. We found that microaggregates within macroaggre-
gates accounted for only 27% of the macroaggregate weight in
CT, while 47% of the macroaggregate weight in NT consists of
microaggregates (Table 1). Hence, the formation of new
microaggregates within macroaggregates was reduced by a
factor of about 2 (27 versus 47%) in CT compared to NT.
These data are congruent with the previous observation that
macroaggregates turn over twice as fast in CT compared to NT
based on the ratio of fine iPOM to coarse iPOM (Fig. 2). Thus,
macroaggregate turnover and microaggregate formation
appear to be linked: a doubling of macroaggregate turnover
reduces the microaggregate formation within macroaggre-
gates by a factor of 2.
Within macroaggregates, the concentration of intra-
J. Six et al. / Soil Biology & Biochemistry 32 (2000) 2099–2103 2101
Fig. 2. Ratio of fine intra-particulate organic matter (iPOM) over coarse
iPOM in the surface layer (0–5 cm) of NT and CT soils across four
agricultural experiment sites (Sidney, NE; Wooster, OH; Kellogg Bio-
logical Station, MI; Lexington, KY). The ratio’s are calculated based on
data presented in Six et al. (1999a).
microaggregate POM-C was 3-fold greater in NT compared
to CT while the concentration of inter-microaggregate
POM-C (i.e. POM-C held within macroaggregates but not
within microaggregates) was 2-fold greater in CT compared
to NT (Table 1). In addition, the proportion of fine iPOM-C
held within microaggregates that occurs within macroaggre-
gates was much larger in NT compared to CT (90 versus
58%, respectively) leading us to conclude that the slower
macroaggregate turnover in NT leads to a sequestration of
C, primarily as fine iPOM held within stable microaggre-
gates that are formed within macroaggregates. The greater
accumulation of microaggregates rich in fine iPOM contri-
butes to the greater total SOM-C under NT. Six et al.
(1999a) found that the increase in total fine iPOM (i.e.
inter- plus intra-microaggregate iPOM) alone accounted,
on average, for 21% of the total C difference between NT
and CT at four agricultural experiment sites.
Both the microaggregates within macroaggregates and
the intra-microaggregate POM-C were younger in NT
than in CT (Table 2). This is attributed to the slower turn-
over of macroaggregates in NT, resulting in more seques-
tration of crop-derived C within stable microaggregates
formed within macroaggregates under NT. In addition, the
d 13C signature of the mineral fraction (,53 mm) contained
within the macroaggregates suggests that C sequestration is
also enhanced in this fraction (Table 2).
The halved rate of macroaggregate turnover, as suggested
by the data in Fig. 2, results in twice as many microaggre-
gates formed within macroaggregates which, in turn, leads
to twice as much fine iPOM accumulating in NT compared
to CT. We have reported a doubling of both the amount of
macroaggregates (Six et al., 2000) and the mean residence
time of total soil C (Paustian et al., 2000; Six et al., 1999a) in
NT compared to CT in several soils.
The linkage of macroaggregate turnover, microaggregate
formation and SOM dynamics hypothesized in our model,
was strongly supported by our data on the composition of
macro- and microaggregates and their respective C
dynamics. The linkage of these processes gives a mechan-
istic explanation as to why there is greater accumulation of
SOM-C under NT agriculture. Our results underscore the
importance of soil aggregation and more specifically the
interactions of SOM and aggregate dynamics in controlling
C-sequestration in soils.
Acknowledgements
We appreciate the assistance of Dan Reuss, Karolien
Denef and Heleen Bossuyt during methods development.
We thank Jason Kaye, Rich Conant, and Serita Frey for
comments on the manuscript. Clay Combrink and Caroline
Krumm are acknowledged for help with analyses. This work
was funded by the US National Science Foundation and the
US Department of Energy.
J. Six et al. / Soil Biology & Biochemistry 32 (2000) 2099–21032102
Table 1
Percent microaggregates in macroaggregates and concentrations of inter-microaggregate particulate organic matter (POM) and intra-microaggregate POM in
no-tillage (NT) and conventional tillage (CT) surface layer (0–5 cm) at Sidney, NE
Treatment Microaggregates in
macroaggregates a
Inter-microaggregate POM Intra-microaggregate POM
(%) (g C kg21 sandfree macroaggregates) (g C kg21 sandfree macroaggregates)
NT 47.1 ab 1.3 b 11.1 a
CT 27.0 b 2.8 a 3.8 b
a The data presents the % of macroaggregate soil found in microaggregates (sand-corrected).
b Values followed by a different lowercase letter between management treatments are significantly different
P , 0:05 according to Tukey’s HSD mean
separation test.
Table 2
d 13C signature (per mille) of different aggregate size classes and particulate organic matter (POM) fractions in the surface layer (0–5 cm) of no-tillage (NT)
and conventional tillage (CT at Sidney, NE). Wheat residues had a d 13C of 227.57 compared to a d 13C of 218.43 for the native grassland soil present prior to
initial cultivation in 1969. Thus more negative values indicate a greater proportion of crop-derived carbon and consequently a younger age
Fraction NT CT
(per mille) (per mille) (per mille)
Macroaggregate (250–2000 mm) 2 21.70 a Ba 2 21.96 a B
Microaggregate within macroaggregate 2 21.39 a B 2 20.32 b C
Intra-microaggregate POM 2 23.56 a A 2 22.73 b A
Mineral fractionb 2 19.89 a D 2 18.47 b E
Free microaggregate (53–250 mm) 2 20.55 a C 2 19.19 b D
a Values followed by a different lowercase letter between management treatments and within a fraction are significantly different
P , 0:05 according to
Tukey’s HSD mean separation test. Values followed by a different uppercase letter between fraction and within a management treatment are significantly
different
P , 0:05 according to Tukey’s HSD mean separation test.
b Macroaggregates contain ,53 mm particles.
References
Angers, D.A., Recous, S., Aita, C., 1997. Fate of carbon and nitrogen in
water-stable aggregates during decomposition of 13C15N-labelled wheat
straw in situ. European Journal of Soil Science 48, 295–300.
Elliott, E.T., Coleman, D.C., 1988. Let the soil work for us. Ecological
Bulletins 39, 23–32.
Gale, J., Cambardella, C.A., Bailey, T.B., 2000. Root-derived carbon and
the formation and stabilization of aggregates. Soil Science Society of
America Journal 64, 201–207.
Golchin, A., Oades, J.M., Skjemstad, J.O., Clarke, P., 1994. Study of free
and occluded particulate organic matter in soils by solid state 13C CP/
MAS NMR spectroscopy and scanning electron microscopy. Australian
Journal of Soil Research 32, 285–309.
Guggenberger, G., Christensen, B.T., Zech, W., 1994. Land-use effects on
the composition of organic matter in particle-size separates of soil: I.
Lignin and carbohydrate signature. European Journal of Soil Science
45, 449–458.
Jastrow, J.D., 1996. Soil aggregate formation and the accrual of particulate
and mineral-associated organic matter. Soil Biology & Biochemistry
28, 656–676.
Jastrow, J.D., Boutton, T.W., Miller, R.M., 1996. Carbon dynamics of
aggregate-associated organic matter estimated by carbon-13 natural
abundance. Soil Science Society of America Journal 60, 801–807.
Oades, J.M., 1984. Soil organic matter and structural stability: mechanisms
and implications for management. Plant and Soil 76, 319–337.
Paul, E.A., Paustian, K., Elliott, E.T., Cole, C.V. (Eds.), 1997. Soil Organic
Matter in Temperate Agroecosystems: Long-term Experiments in North
America CRC Press, Boca Raton.
Paustian, K., Collins, H.P., Paul, E.A., 1997. Management controls on soil
carbon. In: Paul, E.A., Paustian, K., Elliott, E.A., Cole, C.V. (Eds.). Soil
Organic Matter in Temperate Agroecosystems: Long-term Experiments
in North America, CRC Press, Boca Raton, pp. 15–49.
Paustian, K., Six, J., Elliott, E.T., Hunt, H.W., 2000. Management options
for reducing CO2 emissions from agricultural soils. Biogeochemistry
48, 147–163.
Six, J., Elliott, E.T., Paustian, K., Doran, J.W., 1998. Aggregation and soil
organic matter accumulation in cultivated and native grassland soils.
Soil Science Society of America Journal 62, 1367–1377.
Six, J., Elliott, T., Paustian, K., 1999a. Aggregate and soil organic matter
dynamics under conventional and no-tillage systems. Soil Science
Society of America Journal 63, 1350–1358.
Six, J., Schultz, P.A., Jastrow, J.D., Merckx, R., 1999b. Recycling of
sodium polytungstate used in soil organic matter studies. Soil Biology
& Biochemistry 31, 1193–1196.
Six, J., Paustian, K., Elliott, E.T., Combrink, C., 2000. Soil structure and
soil organic matter: I. Distribution of aggregate size classes and aggre-
gate associated carbon. Soil Science Society of America Journal 64,
681–689.
Skjemstad, J.O., Le Feuvre, R.P., Prebble, R.E., 1990. Turnover of soil
organic matter under pasture as determined by 13C natural abundance.
Australian Journal of Soil