Geoderma 138 (2007) 204 – 212
www.elsevier.com/locate/geoderma
Free atmospheric CO2 enrichment (FACE) increased respiration
and humification in the mineral soil of a poplar plantation
Marcel R. Hoosbeek a,⁎, Judith M. Vos a , Marcel B.J. Meinders a ,
Eef J. Velthorst a , Giuseppe E. Scarascia-Mugnozza b
a
Wageningen University, Department of Environmental Sciences, Earth System Science Group, P.O. Box 47, 6700AA Wageningen, The Netherlands
b
University of Tuscia, Department of Forest Environment and Resources, Via S. Camillo De Lellis, 01100 Viterbo, Italy
Received 28 July 2005; received in revised form 22 August 2006; accepted 17 November 2006
Available online 3 January 2007
Abstract
Free atmospheric CO2 enrichment (FACE) studies conducted at the whole-tree and ecosystem scale indicate that there is a marked increase in
primary production, mainly allocated into below-ground biomass. The enhanced carbon transfer to the root system may result in enhanced
rhizodeposition and subsequent transfer to soil C pools. However, the impact of elevated CO2 on soil C contents has yielded variable results. The
fate and function of this extra C into the soil in response to elevated CO2 are not clear. The POPFACE experiment was initiated early 1999 with the
objective to determine the functional responses of a short-rotation poplar plantation to actual and future atmospheric CO2 concentrations. During
the first 2 years of the second rotation (2002–2003), the increase of total soil C% was larger under FACE than under ambient CO2. Chemical
fractionation revealed the presence of more labile soil C under FACE, which is in agreement with the larger input of plant litter and root exudates
under FACE. In order to gain insight into the fate and function of this extra C into the soil and the dynamics of soil C, we incubated soil samples,
measured respiration rates and used a simple soil C model to interpret the results. FACE increased the accumulated 28-day CO2 production and the
initial Cslow pool content (metabolizable plant remains and partly decomposed SOM). FACE also increased the decomposition rates of the
metabolizable C pools (Cfast + Cslow) in the top soil, while for the subsoil the opposite effect was observed. The modeled formation of humified
SOM was also enhanced by FACE. Our results support the terrestrial feedback hypothesis, i.e. an increase of the long-term terrestrial C sink in
response to increasing atmospheric CO2 concentrations.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Soil carbon; Carbon sequestration; Elevated CO2; FACE; Soil respiration; Soil carbon model
1. Introduction
Two recent estimates of the global net terrestrial carbon (C)
sink yielded respectively between 0.5 Pg C year− 1 for the early
1990s (Joos et al., 1999; Houghton, 2003) and 0.7 Pg C year− 1
during the 1990s (Plattner et al., 2002). These and other
estimates led to many research efforts aimed at gaining insight
into the mechanisms responsible for uptake of C by terrestrial
ecosystems. Since forest vegetation and soils together hold
almost half of the C of the Earth's terrestrial ecosystems, the
effects of predicted future atmospheric CO2 concentrations and
nitrogen (N) deposition rates are being investigated at several
⁎ Corresponding author. Tel.: +31 317 484109; fax: +31 317 482419.
E-mail address: [email protected] (M.R. Hoosbeek).
0016-7061/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.geoderma.2006.11.008
forest FACE (Free air CO2 enrichment) experiments (Schlesinger and Lichter, 2001; Norby et al., 2002; Houghton, 2003;
Hoosbeek et al., 2004). FACE technology has the great merit of
not altering the microclimate of the test area and allows to
perform research on global change impacts at the ecosystem
level (Hendrey et al., 1999). Studies conducted at the whole-tree
and ecosystem scale under elevated CO2 indicate that there is a
marked increase in both above and belowground primary production of forest vegetation (Calfapietra et al., 2001; Prentice
et al., 2001; Hamilton et al., 2002; Norby et al., 2002;
Calfapietra et al., 2003; Liberloo et al., 2006). Enhanced C
transfer to the root system may result mainly in enhanced root
respiration or, otherwise, in an increase of root dry matter,
mycorrhizal activity and subsequent transfer of C to soil C
pools. In a review, Zak et al. (2000) examined the effect of
M.R. Hoosbeek et al. / Geoderma 138 (2007) 204–212
elevated CO2 on fine roots and the response of soil microorganisms. With few exceptions, rates of soil and microbial
respiration were more rapid under elevated CO2, indicating that
(1) greater plant growth under elevated CO2 enhanced the
amount of C entering the soil, and (2) additional substrate was
being metabolized by soil microorganisms.
DeLucia et al. (1997), Fox and Comerford (1990) and Jones
et al. (1998) reported increased exudation of low-molecularweight C compounds by roots of plants grown under elevated
atmospheric CO2 concentrations. At the small-pot scale it has
been demonstrated that, compared to ambient CO2 concentrations, elevated CO2 increases the amount of C allocated to the
rhizosphere by enhancing root deposition (Ineson et al., 1996;
Hungate et al., 1997; Cheng and Johnson, 1998). Based on these
studies, Cheng (1999) concluded that total C input to the
rhizosphere is significantly increased when plants are grown in
elevated CO2. However, the fate and function of this extra C
into the soil in response to elevated CO2 are not clear
(Schlesinger and Lichter, 2001; Norby et al., 2002; Lichter
et al., 2005). Depending on plant and soil conditions, the
increased C input may result in an increase, a decrease, or no
effect on SOM decomposition and nutrient mineralization. The
additional C may be utilized by microorganisms and partly
converted into SOM, thereby increasing soil C content.
Alternatively, the extra C may either stimulate SOM decomposition due to a priming effect, or suppress SOM decomposition
resulting from microbial immobilization.
The POPFACE experiment was initiated early 1999 with the
objective to determine the functional responses of a short-rotation
poplar plantation to actual and future atmospheric CO2 concentrations, and to assess the interactive effects of this anthropogenic
perturbation with the other natural environmental constraints on
key biological processes and structures (Scarascia-Mugnozza
et al., 2006). Additionally, this experiment was intended to yield
data relevant to assess the potential for increasing the C
sequestering capacity and the woody biomass production within
the European Union, using such forest tree plantations.
During the first rotation (1999–2001), total soil C content
increased under ambient CO2 and FACE treatment respectively
with 12 and 3%, i.e. 484 and 107 g C/m2 (Hoosbeek et al., 2004).
We estimated the input of new C with the C3/C4 stable isotope
(δ13C) method. Respectively 704 and 926 g C/m2 of new C was
incorporated under control and FACE during the 3 year experiment. Although more new C was incorporated under FACE,
the increase of total C was suppressed under FACE. We hypothesized that these seemingly opposite effects may have been
caused by a priming effect of the newly incorporated litter, where
priming effect is defined as the stimulation of SOM decomposition caused by the addition of labile substrates.
In 2002 the experiment continued with a second three-year
rotation (named EuroFACE). In order to verify our hypothesis
and to gain insight into the effects of FACE, N-fertilization and
poplar species on soil C and N dynamics, we applied chemical
fractionation, measured rates of N-mineralization, and obtained
respiration curves. Sampling for these analyses took place
during the second and third year of the second rotation, i.e. in
2003 and 2004 respectively. In contrast to the first rotation, soil
205
analyses of the first two years of the second rotation showed a
larger increase of total soil C% under FACE than under ambient
CO2 (Hoosbeek et al., 2006). Based on these observations we
inferred that the priming effect ceased during the second
rotation. Nevertheless, chemical fractionation revealed an
increase of the labile C fraction at 0–10 cm depth due to
FACE treatment (Hoosbeek et al., 2006), which is in agreement
with the larger input of plant litter and root exudates under
FACE (Lukac et al., 2003; Liberloo et al., 2006). Nmineralization rates were not affected by FACE. We inferred
that the system underwent transition from a state where extra
labile C and sufficient N-availability (due to the former
agricultural use of the soil) caused a priming effect (first
rotation), to a state where extra C input is accumulating due to
limited N-availability (second rotation).
In this article we report on the respiration measurements and
present a simple soil C model in order to interpret the obtained
respiration curves and to gain insight into the dynamics of soil C
in the incubated samples. Based on previous results, i.e.
increased NPP and the presence of more labile C under FACE,
we hypothesize that FACE will (1) increase CO2 respiration
during incubation, (2) increase the initial C contents of SOM
pools, (3) increase SOM decomposition rates, and (4) possibly
increase humification rates due to increased turnover.
2. Methods
The POPFACE experiment was established early 1999 on
former agricultural fields near Viterbo (42°37′04″N, 11°80′87″
E, alt 150 m), Italy. The plantation and adjacent fields had been
under forest until about 1950. Since then a variety of agricultural
crops has been grown on these former forest soils until the
inception of the POPFACE plantation. The annual precipitation
is on average 700 mm with dry summers (Xeric moisture
regime). The loamy soils were classified as Pachic Xerumbrepts
and were described in detail by Hoosbeek et al. (2004). Soil pH
values ranged between 4.8 and 5.5 among the plots.
Nine hectares were planted with Populus x euramericana
uniform hardwood cuttings (length 25 cm) at a density of 0.5 trees
per m2. Within this plantation three FACE and three control plots
(30 × 30 m2) were randomly assigned under the condition of
minimum CO2 enrichment pollution. The plots were divided into
two parts by a resin-glass barrier (1 m deep in the soil) for nitrogen
differential treatments in the two halves of each plot. However,
because of the high inorganic N content of the soil (Hoosbeek
et al., 2006), no fertilization treatment was applied during the first
3-year rotation of the experiment. Each half plot was divided into
three sectors, where each sector was planted at a density of 1 tree
per m2 using three different genotypes: P. x euramericana Dode
(Guinier) (=P. deltoides Bart. ex Marsh. x P. nigra L.) genotype
I-214, a genotype of P. nigra L. (Jean Pourtet) and a local selection of P. alba L. (genotype 2AS11). Carbon enrichment was
achieved by injection of pure CO2 through laser-drilled holes in
tubing mounted on six masts (Miglietta et al., 2001). The FACE
rings (octagons) within the FACE plots had a diameter of about
22 m. The elevated CO2 concentrations, measured at 1-min
intervals, were within 20% deviation from the pre-set target
206
M.R. Hoosbeek et al. / Geoderma 138 (2007) 204–212
concentration (560 μmol mol− 1) for 91% of the time to 72.2% of
the time, respectively, at the beginning and at the end of each
rotation cycle of the plantation. The plantation was drip irrigated
at a rate of 6 to 10 mm per day during the growing seasons.
The trees were coppiced after the first three growing seasons
(1999–2001). The experiment continued with a second rotation
under the name EUROFACE (2002–2004). A fertilization treatment was added to one half of each experimental plot because soil
analyses showed the occurrence of limiting conditions of nitrogen
availability in the soil (Scarascia-Mugnozza et al., 2006). The total
amount of nitrogen supplied was 212 kg ha− 1 year− 1 in 2002 and
290 kg ha− 1 year− 1 during 2003 and 2004. The nitrogen was
supplied through the irrigation system in constant weekly amounts
with a 4:1 NH4+:NO3− ratio in 2002, while in 2003 and 2004 the
nitrogen was supplied in weekly amounts proportional to the
growth rate with a 1:1 NH4+:NO3− ratio.
Soil samples were collected from each sector within the 3
control and 3 FACE plots in October of 2003 and September of
2004. Bulk density samples were taken with 300 cm3 metal
rings at 0–10 and 10–20 cm below the surface of the mineral
soil. The samples were dried at 105 °C for 3 days. Bulk densities
were calculated based on dry weight and ring volume. Next, the
soil samples were crushed by hand and live roots were removed.
Carbon and nitrogen were determined by flash combustion in an
elemental analyzer (EA 1108) (Van Lagen, 1996). Total soil
organic C and N contents are expressed on dry weight basis (g C
or N per gram dry soil × 100%).
For respiration measurements, field moist bulk samples were
passed through an 8 mm sieve with great care in order to minimize
disruption of the natural aggregate structure, and stored at 4 °C.
Upon incubation, 175 g of field moist soil was put in a 300 ml
vessel and placed in a temperature controlled chamber and
connected to the respirometer (ADC Bioscientific Ltd., England,
24-vessel Gas Handling Unit in combination with a 2250 Gas
analyzer). Subsamples were analyzed for moisture content.
Carbon dioxide-scrubbed air was continuously passed through
the head space of all 24 vessels, while the CO2 concentration in
one of the 24 vessels was measured for 5 min. Towards the end of
each 5-min period, after enough time had passed for flushing and
stabilizing, the CO2 concentration was recorded. This schedule
yielded one record per 2 h per sample. The readings from the IRcell (mg CO2 m− 3) were multiplied by the gas flow (m3 h− 1) in
order to yield respiration rates (mg CO2 h− 1). These rate values
−1
were converted to μg CCO2 h− 1 gsoil
. The obtained respiration
curves were fitted with the following equation:
ð1Þ
where A, a, B, and b are fit parameters. Integration yields the
amount of produced CCO2 over time interval t:
0
t
dCCO2
A
B
dt ¼ − e−ad t − e−bd t
a
b
dt
CCO2 ðt Þ ¼
The Cfast pool is considered to contain easily metabolizable
organic matter (e.g. easily decomposable carbohydrates from
leaf and root litter, root exudates). Due to sample handling this
pool may also contain microbial necromass. The Cslow pool
contains metabolizable organic matter with longer turnover
times (e.g. resistant plant material, partly decomposed organic
matter). From the Cfast pool C is respired to the CCO2 pool and
partly incorporated into the Cslow pool. And from the Cslow pool
C is respired to the CCO2 pool and partly incorporated into the
Chumified pool. Respiration from the Chumified pool was assumed
to be negligible during incubation. This assumption is based on
the relatively long turn over time of humified C as compared to
the length of incubation. The corresponding set of ordinary
differential equation (ODE) in a linear approximation consists of three coupled ODE's (with Cfast = C1, Cslow = C2,
Chumified = C3 and CCO2 = C4)
dC1
¼ −k1 C1 −k12 C1
dt
ð3Þ
dC2
¼ k12 C1 −k2 C2 −k23 C2
dt
ð4Þ
dC3
¼ k23 C2
dt
ð5Þ
and the ODE describing the CO2 flux
dCCO2
¼ Ae−ad t þ Be−bd t
dt
Z
In order to gain insight into the dynamics of organic C in the
incubated sample, we used a simple soil C model with three soil
C pools (Cfast, Cslow, and Chumified).
j
t
0
A
B
ð1−e−ad t Þ þ 1−e−bd t
a
b
ð2Þ
dC4
¼ k1 C 1 þ k2 C 2
dt
ð6Þ
These ODE's can be solved analytically (see Appendix 1). The
contents of the soil C pools as functions of time and the initial
contents of C1 (C01) and C2 (C02) are given in Eqs. (24), (25),
(29), (30), and (31).The decomposition rates of Cfast and Cslow
are represented by respectively a ( = k1 + k12) and b ( = k2 + k23).
The SPSS (v 11.5) General Linear Model was used to
calculate univariate analysis of variance and to evaluate FACE
and N-fertilization treatment and species effects. Differences
M.R. Hoosbeek et al. / Geoderma 138 (2007) 204–212
207
Table 1
Total soil carbon and nitrogen weight percentages (November 2003)
Table 2
1
Accumulated 28-day CCO2 production (μg CCO2 g−soil
)
Soil
depth
(cm)
Soil
depth
Treatment
n
C%
N%
Bulk
density
(g soil cm−3)
Mean S.E. Mean S.E. Mean S.E.
Ambient
FACE
N
Ambient
Fertilized
Species alba
nigra
euramericana
10–20 CO2
Ambient
FACE
N
Ambient
Fertilized
Species alba
nigra
euramericana
0–10
CO2
18
18
18
18
12
12
12
18
18
18
18
12
12
12
1.22
1.11
1.18
1.15
1.15
1.15
1.19
1.04
0.90
0.97
0.96
0.96
0.98
0.97
0.03
0.04
0.04
0.03
0.04
0.05
0.05
0.01
0.04
0.04
0.04
0.04
0.05
0.05
0.12
0.11
0.11
0.11
0.11
0.11
0.12
0.10
0.09
0.10
0.10
0.10
0.10
0.10
0.00
0.01
0.01
0.01
0.01
0.01
0.01
0.00
0.01
0.01
0.01
0.01
0.01
0.01
1.29
1.20
1.24
1.25
1.24
1.24
1.26
1.31
1.25
1.26
1.30
1.27
1.31
1.27
0.02
0.03
0.03
0.03
0.03
0.04
0.03
0.02
0.02
0.02
0.02
0.03
0.03
0.02
between means were considered significant when the P-value
of the UNIANOVA F-test was b0.05.
3. Results
The total soil carbon percentages (Ctotal) of the top 10 cm of
the mineral soil were respectively 1.22 and 1.11% in the
1
Fig. 1. (a and b) Average respiration rates (μg CCO2 h− 1 g−soil
) (ambient CO2 vs.
FACE) of soil samples taken at 0–10 cm depth for 2003 (a) and 2004 (b). For
statistics on respiration, see Table 2.
Treatment
(cm)
0–10
CO2
N
Species
10–20
CO2
N
Species
Ambient
FACE
Ambient
Fertilized
alba
nigra
euramericana
Ambient
FACE
Ambient
Fertilized
alba
nigra
euramericana
2003
2004
CCO2 respiration
CCO2 respiration
n
Mean
S.E. n
Mean
S.E.
54
54
54
54
36
36
36
54
54
54
54
36
36
36
57.6
60.1
58.5
59.3
53.0
57.8
65.9
37.3
37.8
36.5
38.6
36.0
42.1
34.9
3.5
4.1
3.3
4.3
4.1
4.6
5.1
3.7
2.1
3.0
3.1
3.8
4.1
3.4
52.6
69.4
65.6
56.1
50.7
77.9
53.8
29.8
43.4
37.2
35.0
34.6
41.8
32.8
3.6
4.8
4.4
4.3
3.7
6.5
3.9
2.8
4.7
4.8
3.3
4.6
7.1
3.0
36
36
36
36
24
24
24
18
18
18
18
12
12
12
a
b
c
a. Significant CO2 treatment effect (P = 0.002) on CCO2 produced in 28 days
(2004) at depth 0–10 cm.
b. Significant species effect (P = 0.000) on CCO2 produced in 28 days (2004) at
depth 0–10 cm.
c. Significant CO2 treatment effect (P = 0.010) on CCO2 produced in 28 days
(2004) at depth 10–20 cm.
ambient and FACE plots (Table 1). Between 10 and 20 cm soil
depth, Ctotal was respectively 0.90 and 1.04% in the ambient and
FACE plots. Average total N percentages (Ntotal) at depths 0–10
and 10–20 cm were respectively 0.11 and 0.10%. Average bulk
densities at both depths were almost similar, i.e. 1.23 and 1.26 g
soil cm− 3.
All obtained respiration curves consisted of two distinct parts
(Fig. 1a and b). A “steep” part during approximately the first 24 h
with sharply decreasing respiration rates, followed by a part with
gradually decreasing rates. Soil samples from FACE plots had
higher respiration rates than those from ambient plots for about
the first 10 days (2003 and 2004). After approximately the first
10 days, respiration rates of FACE samples became lower than
those of ambient samples. FACE increased the accumulated 28day CCO2 production at both depths, but significantly only in
2004 (Table 2). N treatment did not affect the 28-day CCO2
production. In 2004, the 28-day CCO2 production was significantly increased by P. nigra at 0–10 cm. The latter trend was
also observed at 10–20 cm, but was not significant.
Initial Cfast content of 2003 was not affected by FACE, N
treatment or species at both depths (Table 3). Initial Cslow
content of 2003 was not affected by FACE, N treatment or
species either, but was about 19 times larger than initial Cfast at
0–10 cm depth and about 9 times larger at 10–20 cm depth. In
2004 there were significant N treatment and species effects on
initial Cfast content of both depths. N-fertilization reduced initial
Cfast content, while P. nigra increased initial Cfast content at
depth 0–10 cm, and P. alba yielded the lowest initial Cfast
content at both depths. The initial Cslow content of 2004 was
significantly increased by FACE at both depths. P. nigra
increased initial Cslow content significantly at depth 0–10 cm.
The decomposition rates of Cfast were on average 24 (2003)
to 29 (2004) times faster than of Cslow at depth 0–10 cm, and 27
208
M.R. Hoosbeek et al. / Geoderma 138 (2007) 204–212
Table 3
1
Initial C contents of the Cfast and Cslow pools (μg C g−soil
)
Soil
depth
Treatment
2003
Initial Cfast
Mean
S.E.
Mean
S.E.
54
54
54
54
36
36
36
54
54
54
54
36
36
36
6.59
6.22
6.42
6.39
5.64
6.21
7.36
7.69
7.92
8.52
7.06
7.17
8.85
7.48
0.43
0.59
0.44
0.58
0.39
0.70
0.72
1.05
0.94
1.07
0.91
1.11
1.10
1.42
125.14
119.42
118.24
126.19
109.60
125.99
131.18
69.04
70.68
68.60
71.06
67.05
74.43
68.47
8.76
8.69
7.13
10.03
9.78
11.43
10.63
7.61
4.45
6.66
5.94
7.19
7.71
8.35
(cm)
0–10
CO2
N
Species
10–20
CO2
N
Species
Ambient
FACE
Ambient
Fertilized
alba
nigra
euramericana
Ambient
FACE
Ambient
Fertilized
alba
nigra
euramericana
2004
Initial Cslow
n
n
Initial Cfast
Initial Cslow
Mean
S.E.
Mean
36
36
36
36
24
24
24
18
18
18
18
12
12
12
10.55
12.00
13.21
9.25
9.54
13.93
10.27
6.60
6.51
7.38
5.63
4.57
6.40
8.34
0.98
1.03
1.17
0.66
1.04
1.57
0.78
1.02
0.78
1.06
0.64
0.52
1.25
1.16
98.81
123.70
117.10
104.89
92.31
143.61
96.73
50.17
94.80
73.09
68.82
76.32
79.53
59.70
a
b
c
d
S.E.
e
f
g
7.01
9.25
8.25
8.53
7.14
12.20
7.37
5.35
11.45
10.99
9.39
13.76
14.15
9.74
a. Significant N treatment effect (P = 0.002) on initial Cfast (2004) at depth 0–10 cm.
b. Significant species effect (P = 0.009) on initial Cfast (2004) at depth 0–10 cm.
c. Significant N treatment effect (P = 0.021) on initial Cfast (2004) at depth 10–20 cm.
d. Significant species effect (P = 0.010) on initial Cfast (2004) at depth 10–20 cm.
e. Significant CO2 treatment effect (P = 0.017) on initial Cslow (2004) at depth 0–10 cm.
f. Significant species effect (P = 0.000) on initial Cslow (2004) at depth 0–10 cm.
g. Significant CO2 treatment effect (P = 0.001) on initial Cslow (2004) at depth 10–20 cm.
(2003) to 24 (2004) times faster at depth 10–20 cm (Table 4).
During 2003, FACE significantly increased decomposition rates
of Cfast and Cslow at 0–10 cm depth. However, at 10–20 m
depth, FACE significantly decreased the decomposition rate of
Cslow. The same opposite FACE effect for each depth was
observed for 2004. Again, FACE significantly increased decomposition rates of Cfast and Cslow at 0–10 cm depth, while at
10–20 cm FACE significantly decreased decomposition rates of
both Cfast and Cslow. No N treatment effects were observed for
2003 and 2004. The only significant species effect was
Table 4
1
) of the Cfast and Cslow pools
Decomposition rates (μg CCO2 h− 1 g−soil
Soil
depth
Treatment
2003
(cm)
0–10
CO2
N
Species
10–20
2004
Cslow
Cfast
CO2
N
Species
Ambient
FACE
Ambient
Fertilized
alba
nigra
euramericana
Ambient
FACE
Ambient
Fertilized
alba
nigra
euramericana
Cfast
Cslow
n
Mean
S.E.
Mean
S.E.
n
Mean
S.E.
Mean
S.E.
54
54
54
54
36
36
36
54
54
54
54
36
36
36
0.099
0.124
0.114
0.109
0.106
0.110
0.118
0.180
0.180
0.164
0.196
0.179
0.188
0.172
0.006
0.006
0.007
0.006
0.007
0.008
0.008
0.018
0.012
0.013
0.017
0.016
0.023
0.017
0.0037
0.0055
0.0047
0.0046
0.0046
0.0043
0.0050
0.0072
0.0059
0.0066
0.0066
0.0067
0.0067
0.0064
0.0002
0.0002
0.0002
0.0002
0.0003
0.0003
0.0003
0.0005
0.0003
0.0005
0.0003
0.0005
0.0005
0.0005
36
36
36
36
24
24
24
18
18
18
18
12
12
12
0.108
0.210
0.165
0.152
0.155
0.151
0.170
0.236
0.072
0.154
0.165
0.179
0.141
0.158
0.011
0.009
0.011
0.015
0.020
0.013
0.015
0.012
0.004
0.022
0.022
0.033
0.023
0.025
0.0040
0.0069
0.0056
0.0053
0.0056
0.0049
0.0059
0.0095
0.0035
0.0066
0.0068
0.0063
0.0060
0.0076
0.0002
0.0003
0.0003
0.0003
0.0004
0.0003
0.0004
0.0005
0.0003
0.0009
0.0008
0.0012
0.0008
0.0010
a
b
c
a. Significant CO2 treatment effect (P = 0.005) on Cfast (2003) decomposition rate at depth 0–10 cm.
b. Significant CO2 treatment effect (P = 0.000) on Cslow (2003) decomposition rate at depth 0–10 cm.
c. Significant CO2 treatment effect (P = 0.039) on Cslow (2003) decomposition rate at depth 10–20 cm.
d. Significant CO2 treatment effect (P = 0.000) on Cfast (2004) decomposition rate at depth 0–10 cm.
e. Significant CO2 treatment effect (P = 0.000) on Cfast (2004) decomposition rate at depth 10–20 cm.
f. Significant CO2 treatment effect (P = 0.000) on Cslow (2004) decomposition rate at depth 0–10 cm.
g. Significant species effect (P = 0.026) on Cslow (2004) decomposition rate at depth 0–10 cm.
h. Significant CO2 treatment effect (P = 0.000) on Cslow (2004) decomposition rate at depth 10–20 cm.
i. Significant species effect (P = 0.004) on Cslow (2004) decomposition rate at depth 10–20 cm.
d
e
f
g
h
i
M.R. Hoosbeek et al. / Geoderma 138 (2007) 204–212
Table 5
1
Accumulated 28-day formation of Chumified (μg Chumified g−soil
)
Soil
depth
Treatment
(cm)
0–10
CO2
N
Species
10–20
CO2
N
Species
Ambient
FACE
Ambient
Fertilized
alba
nigra
euramericana
Ambient
FACE
Ambient
Fertilized
alba
nigra
euramericana
2003
2004
Chumified (672 h)
Chumified (672 h)
n
Mean S.E. n
Mean
S.E.
54
54
54
54
36
36
36
54
54
54
54
36
36
36
54.3
57.0
55.3
56.1
50.1
54.7
62.2
33.4
33.9
32.2
35.1
32.5
37.7
31.2
47.4
63.4
59.0
51.5
46.0
70.9
48.6
26.5
40.1
33.5
32.2
32.3
38.6
28.6
3.2
4.5
4.0
4.1
3.5
6.0
3.6
2.7
4.4
4.4
3.3
4.5
6.5
2.8
3.4
3.8
3.1
4.1
4.0
4.3
4.8
3.3
1.8
2.6
2.8
3.3
3.8
2.8
36
36
36
36
24
24
24
18
18
18
18
12
12
12
a
b
c
a. Significant CO2 treatment effect (P = 0.001) on humification (Chumified at
t = 672 h; 2004) at depth 0–10 cm.
b. Significant species effect (P = 0.000) on humification (Chumified at t = 672 h;
2004) at depth 0–10 cm.
c. Significant CO2 treatment effect (P = 0.008) on humification (Chumified at
t = 672 h; 2004) at depth 10–20 cm.
observed in 2004 for Cslow. At depth 0–10 cm, P. nigra induced
lower Cslow decomposition rates (as compared to both other
species), while at 10–20 cm P. euramericana increased Cslow
decomposition rates.
FACE treatment significantly increased the 28-day formation
of Chumified at both depths in 2004 (Table 5). N treatment did not
affect the formation of Chumified. P. nigra increased the
formation of Chumified in the top soil during 2004.
4. Discussion
Total soil C% has been increasing in all plots, i.e. under all
treatments and poplar species, since the beginning of the
experiment. This increase in C% is largely due to the
afforestation of agricultural land. During the first rotation
(1999–2001), total soil C content increased more under ambient
CO2 treatment than under FACE, while under FACE more new
C was incorporated than under ambient CO2 (Hoosbeek et al.,
2004). These unexpected and opposite effects may have been
caused by a priming effect. The extra available C under FACE
probably increased the decomposition of old and new soil C. In
order to verify our hypothesis, we measured total soil C%,
applied chemical fractionation, measured rates of N-mineralization and obtained respiration curves based on samples
collected during the second rotation (2002–2004).
Results of the first two years of the second rotation show a
larger increase of total C% under FACE than under ambient
CO2 (Hoosbeek et al., 2006). In contrast to the first rotation,
SOM is now accumulating faster under FACE than under
ambient CO2. Based on these observations we concluded that
the priming effect ceased during the second rotation. Although
the priming effect ceased, chemical fractionation revealed a
continued presence of more labile C under FACE. This is in
209
agreement with the larger input of plant litter and root exudates
under FACE (Lukac et al., 2003; Liberloo et al., 2006). We
therefore hypothesized that FACE would (1) increase CO2
respiration, (2) increase initial pool sizes, (3) increase
decomposition rates, and (4) possibly increase humification
rates due to increased turnover. FACE indeed increased the CO2
respiration significantly in 2004 at both depths, while the same
trend was observed for 2003 (Table 2).
The initial Cfast content was not affected by FACE (Table 3).
This may in part be due to the very labile nature of Cfast. The
turnover times of Cfast, i.e. the inverse of the decomposition
rates (Table 4), range between 5 and 10 h at 0–10 cm depth in
2003 and 2004. N-fertilization did, however, significantly
decrease the initial Cfast content at both depths in 2004, while
the same trend was observed for 2003. Although the C pools
cannot be compared directly, the labile C pool obtained by
chemical fractionation was also decreased by N-fertilization
(Hoosbeek et al., 2006). The observed N-fertilization effect
on initial Cfast confirms our conclusion that microbial growth is
N-limited. As opposed to the first rotation, limited N-availability during the second rotation in the unfertilized plots may
have limited microbial growth which may have allowed a partial
accumulation of new, less labile, soil C (i.e. largely Cslow).
The initial Cslow pool made up 96 and 92% of the
metabolizable pools (Cfast + Cslow) in respectively 2003 and
2004 at a depth of 0–10 cm. Cslow is considered to consist
largely of metabolizable plant remains and partly decomposed
SOM (without the “fast” sugar-like components). In contrast to
the initial Cfast pool, FACE did significantly increase the initial
Cslow contents in 2004 at both depths, which is in support of our
hypothesis and is in agreement with the observed increased root
turnover under FACE (Lukac et al., 2003).
Poplar species had some effect on initial Cfast and Cslow
contents, i.e. in 2004 P. nigra increased Cfast and Cslow at 0–
10 cm depth, while alba decreased Cfast at 10–20 cm. These
effects may have been a combination of increased root exudation
and/or increased root turnover. Although, during the first
rotation Lukac et al. (2003) found alba to have the fastest root
turnover rates (1.6 vs. 2.3 year− 1 for ambient vs. FACE), followed by nigra (1.6 vs. 2.0 year− 1), and euramericana (1.4 vs.
1.8 year − 1 ). In case these root turnover rates also hold for the
second rotation, or at least the relative differences between
the species, they do not support the observed species effects on respiration rates. The observed species effect on
respiration rates may instead possibly be explained by differences in root exudation or mycorrhizal activity between
species.
Although the initial Cfast content was not affected, FACE did
significantly increase decomposition rates at 0–10 cm in 2003
and 2004 (Table 4). This may suggest that, on average,
components of Cfast under FACE were more labile than under
ambient CO2. However, due to the very labile nature of Cfast in
general, and the used sampling and respiration method, results
with respect to Cfast, i.e. primarily the first part of the curves,
need to be interpreted with care. Decomposition rates of Cslow in
the top soil were significantly enhanced by FACE in 2003 and
2004. This suggests that either more Cslow was present under
210
M.R. Hoosbeek et al. / Geoderma 138 (2007) 204–212
FACE, which is certainly true for 2004, or that the Cslow fraction
consisted of relatively more labile organic matter. At the second
depth, however, FACE had an opposite effect on decomposition
rates of the Cslow fraction. A possible explanation may be found
in the change of type and function of roots with depth. We
observed the ratio fine to coarse root biomass to be larger in the
top 10 cm as compared to the 10–20 cm depth increment. In the
top 10 cm, rhizodeposition may primarily consist of exudation
of carbohydrates by fine roots and the turnover of fine roots. At
10–20 cm depth, the turnover of relatively coarser roots will
make up a larger part of the rhizodeposition. In the top 10 cm,
FACE probably increased primarily root exudation and fine root
turnover, inducing less stable C input with faster decomposition
rates. At 10–20 cm depth, however, FACE probably increased
relatively more coarse root biomass, which resulted in
rhizodeposition with relatively lower decomposition rates.
The fate of extra new C in the soil in response to FACE
depends on how much is utilized by microorganisms and partly
converted into humified SOM (Cheng, 1999). From the
observed increased respiration rates we infer that FACE
increased soil microbial activity. In our soil carbon model this
is represented by an increase of all flows, including the flow of
C into the Chumified pool (Table 5). The increase of the Chumified
pool under FACE is in agreement with the observed larger
increase of total soil C% under FACE (Hoosbeek et al., 2006).
Until recently, it was reported that increased atmospheric
CO2 concentration did not significantly increase mineral soil C
sequestration at forest FACE experiments (Schlesinger and
Lichter, 2001; Norby et al., 2002; Houghton, 2003; Lichter
et al., 2005). After 6 years of CO2 enrichment at the Duke Forest
FACE experiment, Lichter et al. (2005) did not detect
significant FACE effects on C contents of the mineral soils.
However, the C content of the mineral top soil (0–15 cm)
averaged over the FACE and control rings significantly
increased during the experiment. Physical fractionation suggested that this increase occurred entirely within the free light
fraction in which SOM is not protected against decomposition.
Fractions in which SOM is protected to some degree, i.e. coarse
and fine intra-aggregate particulate organic matter (iPOM) and
mineral associated organic matter were not affected by FACE.
Lichter et al. (2005) concluded that forest soils are unlikely to
sequester significant additional quantities of atmospheric C
associated with CO2 fertilization because of the low rates of C
input to refractory and protected SOM pools.
Recently, Jastrow et al. (2005) raised the question whether the
lack of a FACE effect on soil C content is a general response or a
function of (1) the low statistical power of most experiments, and/
or (2) the magnitude of CO2-stimulated C inputs relative to the
duration of the experiments. At the Oak Ridge deciduous forest
FACE experiment (Norby et al., 2002), organic C in the surface
5 cm of the soil increased linearly during 5 years of CO2
enrichment, while C in the ambient plots remained relatively
constant (Jastrow et al., 2005). A significant FACE effect on soil
C was observed for the top 5 cm. Sampling of a thicker soil
increment, e.g. 0–15 cm, would have “diluted” the increase of C
which would have resulted in a non-significant effect. A metaanalysis of 35 independent experimental observations from a
wide range of ecosystems showed that CO2 enrichment increased
soil C by 5.6% (Jastrow et al., 2005). According to Jastrow et al.
(2005), this result supports the generality of the observed increase
of soil C under FACE at the Oak Ridge experiment.
At the EuroFACE site, we also observed a significant larger
increase in total soil C% after 5 years of CO2 enrichment
(Hoosbeek et al., 2006). Moreover, our respiration measurement
and modeling results show that FACE not only increased
metabolizable soil C pools, but also increased the humified soil
C pool. These results support the terrestrial feedback hypothesis, i.e. an increase of the long-term terrestrial C sink in
response to increasing atmospheric CO2 concentrations.
Acknowledgements
Dr. Carlo Calfapietra and colleagues are gratefully acknowledged for managing and maintaining the FACE facility.
Funding was provided by the European Commission Fifth
Framework Program, Environment and Climate RTD Program,
research contract EVR1-CT-2002-40027 (EUROFACE) and by
the Centre of Excellence “Forest and Climate” of the Italian
Ministry of University and Research (MIUR).
Appendix A
Analytical solution to the coupled ODE's (with Cfast = C1,
Cslow = C2, Chumified = C3 and CCO2 = C4).
The set of coupled ODE's describing the soil C pools is
given in Eqs. (3), (4), and (5).
These do not depend on C4 and can therefore be solved
independently of C4. From the solution of C1 and C2, the CO2
flux is given by Eq. (6).
In matrix notation the coupled set (3, 4, 5) reads
dC
¼ KC
dt
ð7Þ
with vector
0 1
C1
C ¼ @ C2 A
C3
and matrix
0
−ðk1 þ k12 Þ
K¼@
k12
0
ð8Þ
0
−ðk1 þ k12 Þ
k23
1
0
0A
0
ð9Þ
This coupled set of ODE's can be analytically solved. We
sought solutions in the form
C ¼ xekt
ð10Þ
with
0
1
x1
x ¼ @ x2 A
x3
ð11Þ
M.R. Hoosbeek et al. / Geoderma 138 (2007) 204–212
Substituting the trial function (Eq. (10)) into Eq. (7), and
eliminating the common factor exp(kt) yielded the following
eigenvalue problem
Kx ¼ kx
a2 ¼ C01
k12 k23
aða−bÞ
211
ð22Þ
ð12Þ
k23
k12
C02 þ C01
a3 ¼ −
b
ða−bÞ
ð13Þ
and for the contents of the soil C pools as function of time
ð23Þ
or
ðK−kIÞx ¼ 0
with identity matrix I. This equation has non-trivial solutions
when the determinant of K − λI is zero, giving
C1 ðtÞ ¼ C01 e−at
j
C2 ðt Þ ¼ −C01
−a−k
k12
0
0
−b−k
k23
j
0
0 ¼0
−k
ð14Þ
where we substituted
ð24Þ
k12 −at
k12
e þ C02 þ
e−bt
ða−bÞ
ða−bÞ
C3 ðtÞ ¼ a1 þ a2 e−at þ a3 e−bt
ð25Þ
ð26Þ
k1 þ k12 ¼ a
k2 þ k23 ¼ b
ð15Þ
giving
ð−a−kÞð−b−kÞð−kÞ ¼ 0
ð16Þ
Solving this third order polynomial gave three eigenvalues
k1 ¼ 0
The CO2 flux is now given by
dC4
k12 −at
k12 −bt
¼ k1 C01 −k2 C02
e þ k2 C02 þ k2 C01
e
dt
a−b
a−b
−at
−bt
¼ Ae þ Be
where
A ¼ k1 C01 −k2 C01
k12
a−b
ð27Þ
k2 ¼ −a
k3 ¼ −b
ð17Þ
The corresponding eigenvalues were calculated by substituting the eigenvalues into the eigenvalue problem and solving
for x. This gave
0
1
0
1
aða−bÞ
0 1
0
0
B k12 k23 C
B b C
B
a C
x1 ¼ @ 0 A
x2 ¼ B
x3 ¼ @ −
A ð18Þ
C
−
k23
@
A
1
k23
1
1
The general solution of the problem is given by
C ¼ a1 x1 ek1 t þ a2 x2 ek2 t þ a3 x3 ek3 t
ð19Þ
The constant αi was found after defining the boundary
conditions. We defined at t = 0 the initial contents of the soil C
pools to be equal to
0
1
C01
ð20Þ
C0 ¼ @ C02 A
C03
After some algebra we found
k12 k23 1 1
k23
−
a1 ¼ C03 −C01
þ C02
ða−bÞ a b
b
B ¼ k2 C02 þ k2 C01
k12
a−b
ð28Þ
In case we assume respiration and assimilation to be equal,
i.e. k1 = k12 = a/2 and k2 = k23 = b/2, then the initial contents of C1
and C2 can be estimated from the fit parameters
C01 ¼
2A
ab
a− 2ða−bÞ
ð29Þ
C02 ¼
2B
2A
−
b 2a−3b
ð30Þ
and the content of C3 at time t is
C3 ðt Þ ¼ C03
þ
C02 C01 C01 b
C02 C01 a
þ
þ
þ
e−at −
e−bt
2
4
4 a−b
2
4 a−b
ð31Þ
References
ð21Þ
Calfapietra, C., et al., 2001. Growth performance of Populus exposed to “FreeAir Carbon dioxide Enrichment” during the first growing season in the
POPFACE experiment. Annals of Forest Science 58, 819–828.
212
M.R. Hoosbeek et al. / Geoderma 138 (2007) 204–212
Calfapietra, C., et al., 2003. Free-air CO2 enrichment (FACE) enhances biomass
production in a short-rotation poplar plantation. Tree Physiology 23, 805–814.
Cheng, W., 1999. Rhizosphere feedbacks in elevated CO2. Tree Physiology 19,
313–320.
Cheng, W., Johnson, D.W., 1998. Elevated CO2, rhizosphere processes, and soil
organic matter decomposition. Plant and Soil 202, 167–174.
DeLucia, E.H., Callaway, R.M., Thomas, E.M., Schlesinger, W.H., 1997.
Mechanisms of phosphorus acquisition for ponderosa pine under high CO2
and temperature. Annals of Botany 79, 111–120.
Fox, T.R., Comerford, N.B., 1990. Low-molecular-weight organic acids in
selected forest soils of the southeastern USA. Soil Science Society of
America Journal 54, 1139–1144.
Hamilton, J.G., et al., 2002. Forest carbon balance under elevated CO2.
Oecologia 131, 250–260.
Hendrey, G.H., Ellsworth, D.S., Lewin, K.F., Nagy, J., 1999. A free-air enrichment system for exposing tall forest vegetation to elevated atmospheric CO2.
Global Change Biology 5, 293–309.
Hoosbeek, M.R., et al., 2004. More new carbon in the mineral soil of a poplar
plantation under Free Air Carbon Enrichment (POPFACE): cause of
increased priming effect? Global Biogeochemical Cycles 18, GB1040.
Hoosbeek, M.R., Li, Y., Scarascia-Mugnozza, G., 2006. Free atmospheric CO2
enrichment (FACE) increased labile and total carbon in the mineral soil of a
short rotation Poplar plantation. Plant and Soil 281 (1–2), 247–254.
Houghton, R.A., 2003. The contemporary carbon cycle. In: Schlesinger, W.H.
(Ed.), Biogeochemistry. Elsevier, pp. 473–513.
Hungate, B.A., et al., 1997. The fate of carbon in grasslands under carbon
dioxide enrichment. Nature 388, 576–579.
Ineson, P., Cotrufo, M.F., Bol, R., Harkness, D.D., Blum, H., 1996. Quantification of soil carbon inputs under elevated CO2: C3 plants in a C4 soil.
Plant and Soil 187 (2), 345–350.
Jastrow, J.D., et al., 2005. Elevated atmospheric carbon dioxide increases soil
carbon. Global Change Biology 11, 2057–2064.
Jones, T.H., et al., 1998. Impacts of rising atmospheric carbon dioxide on model
terrestrial ecosystems. Science 280, 441–443.
Joos, F., Meyer, R., Bruno, M., Leuenberger, M., 1999. The variability in the
carbon sinks as reconstructed for the last 1000 years. Geophysical Research
Letters 26, 1437–1440.
Liberloo, M., et al., 2006. Woody biomass production during the second rotation
of a bio-energy Populus plantation increases in a future high CO2 world.
Global Change Biology 12 (6), 1094–1106.
Lichter, J., et al., 2005. Soil carbon sequestration and turnover in a pine forest
after six years of atmospheric CO2 enrichment. Ecology 86 (7), 1835–1847.
Lukac, M., Calfapietra, C., Godbold, D.L., 2003. Production, turnover and
mycorrhizal colonization of root systems of three Populus species grown
under elevated CO2 (POPFACE). Global Change Biology 9, 838–848.
Miglietta, F., et al., 2001. Free-air CO2 enrichment (FACE) of a poplar plantation:
the POPFACE fumigation system. New Phytologist 150, 465–476.
Norby, R.J., et al., 2002. Net primary productivity of a CO2-enriched deciduous
forest and the implications for carbon storage. Ecological Applications 12
(5), 1261–1266.
Plattner, G.-K., Joos, F., Stocker, T.F., 2002. Revision of the global carbon
budget due to changing air-sea oxygen fluxes. Global Biogeochemical
Cycles 16 (4), 1096 doi:10.1029/2001GB001746.
Prentice, I.C., et al., 2001. The carbon cycle and atmospheric carbon dioxide. In:
IPCC (Ed.), Climate Change 2001: The Scientific Basis. Cambridge
University Press, Cambridge, pp. 183–237.
Scarascia-Mugnozza, G.E., et al., 2006. Responses to elevated [CO2] of a short
rotation, multispecies poplar plantation: the POPFACE/EUROFACE
experiment. In: Nösberger, J., et al. (Ed.), Managed Ecosystems and CO2.
Ecological Studies. Springer Verlag, Berlin, pp. 173–195.
Schlesinger, W.H., Lichter, J., 2001. Limited carbon storage in soil and litter of
experimental forest plots under increased atmospheric CO2. Nature 411,
466–469.
Van Lagen, B., 1996. Soil analyses. In: Buurman, P., Van Lagen, B., Velthorst, E.J.
(Eds.), Manual for Soil and Water Analyses. Backhuys Publishers, Leiden.
Zak, D.R., Pregitzer, K.S., King, J.S., Holmes, W.E., 2000. Elevated
atmospheric CO2, fine roots and the response of soil microorganisms: a
review and hypothesis. New Phytologist 147, 201–222.