effects of straw carbon input on carbon dynamics in agricultural soils: a meta-analysis
Post on 11-Apr-2017
214 Views
Preview:
TRANSCRIPT
RE S EARCH REV I EW
Effects of straw carbon input on carbon dynamics inagricultural soils: a meta-analysisCHANG L IU * , MENG LU* , J UN CU I , BO L I and CHANGMING FANG
Coastal Ecosystems Research Station of Yangtze River Estuary, Ministry of Education Key Laboratory for Biodiversity and
Ecological Engineering, Institute of Biodiversity Science, Fudan University, Shanghai 200433, China
Abstract
Straw return has been widely recommended as an environmentally friendly practice to manage carbon (C) sequestra-
tion in agricultural ecosystems. However, the overall trend and magnitude of changes in soil C in response to straw
return remain uncertain. In this meta-analysis, we calculated the response ratios of soil organic C (SOC) concentra-
tions, greenhouse gases (GHGs) emission, nutrient contents and other important soil properties to straw addition in
176 published field studies. Our results indicated that straw return significantly increased SOC concentration by
12.8 � 0.4% on average, with a 27.4 � 1.4% to 56.6 � 1.8% increase in soil active C fraction. CO2 emission increased
in both upland (27.8 � 2.0%) and paddy systems (51.0 � 2.0%), while CH4 emission increased by 110.7 � 1.2% only
in rice paddies. N2O emission has declined by 15.2 � 1.1% in paddy soils but increased by 8.3 � 2.5% in upland soils.
Responses of macro-aggregates and crop yield to straw return showed positively linear with increasing SOC concen-
tration. Straw-C input rate and clay content significantly affected the response of SOC. A significant positive relation-
ship was found between annual SOC sequestered and duration, suggesting that soil C saturation would occur after
12 years under straw return. Overall, straw return was an effective means to improve SOC accumulation, soil quality,
and crop yield. Straw return-induced improvement of soil nutrient availability may favor crop growth, which can in
turn increase ecosystem C input. Meanwhile, the analysis on net global warming potential (GWP) balance suggested
that straw return increased C sink in upland soils but increased C source in paddy soils due to enhanced CH4 emis-
sion. Our meta-analysis suggested that future agro-ecosystem models and cropland management should differentiate
the effects of straw return on ecosystem C budget in upland and paddy soils.
Keywords: straw return, soil organic carbon, GHGs emission, carbon sequestration, meta-analysis
Received 10 May 2013 and accepted 6 December 2013
Introduction
Continuous increase in greenhouse gases (GHGs) in the
atmosphere is expected to contribute to global warming
(Lal & Kimble, 1997; Smith & Fang, 2010). How to
reduce and/or mitigate GHGs emission has become an
important global issue (Paustian et al., 2000). It has been
proposed that agricultural soils provide a potential sink
for atmospheric CO2 by sequestrating SOC (Smith et al.,
2000; Pan et al., 2003). By adoption of recommended
management practices (RMPs) in cropping systems, the
annual rate of SOC sequestration was estimated to be
0.4–0.8 Pg C globally, which accounted for 33.3–100%of the total potential of soil C sequestration in the world
(Lal, 2004a). Among all RMPs, straw return has been
suggested as a promising method to increase SOC
sequestration in croplands (Lugato et al., 2006; Lu et al.,
2009). To meet the challenges of managing agricultural
soils to capture atmospheric C, it is important to under-
stand the response of soil C cycle to straw input in
croplands.
Previous work has reported that SOC concentration
(Duiker & Lal, 1999; Li et al., 2010), active SOC fractions
(Chen et al., 2009; Malhi et al., 2011) and three major
biogenic GHGs (i.e., CO2, CH4, and N2O) emission
(Duiker & Lal, 2000; Zou et al., 2005; Ma et al., 2008) can
all be increased by straw incorporation. However,
many uncertainties still remain as to how cropland soils
respond to straw-C input (West & Marland, 2002). For
instance, some studies have shown that straw return
has negligible or negative effects on soil C stocks
(Campbell et al., 1998; Garcia-Orenes et al., 2009). Spe-
cifically, Wang et al. (2011) reported that straw incorpo-
ration to a depth of 0.23 m under conventional tillage
decreased SOC levels in surface layer (0–0.2 m depth).
Niu et al. (2011) also found that straw application stim-
ulated soil C loss at a depth of 0.2–0.4 m. Thus, differ-
ent results among experimental studies have indicated
that a comprehensive understanding of cropping sys-
tem response to straw input remains elusive.
Correspondence: Changming Fang, tel./fax + 86 21 5566 4990,
e-mail: cmfang@fudan.edu.cn
*Contributed equally to this paper.
© 2014 John Wiley & Sons Ltd1366
Global Change Biology (2014) 20, 1366–1381, doi: 10.1111/gcb.12517
Global Change Biology
Straw return does not only directly increase C input
into the soil, but also influences soil physical and chemi-
cal properties and crop growth that determine whether
the benefits of agricultural SOC sequestration can be
realized or not (Lal, 2008). For example, straw return
influences soil nutrient contents. The total amount of
crop residues produced in 2001 was estimated to be
4 9 1010 Mg worldwide, containing about 8 9 107 Mg
of N, P, and K (Lal, 2008). Although the nutrients are
only a small portion of straw, they can be a favorable
complementarity for the soil and provide a long-term
supply of nutrients for crop growth. Returning straw
into soil can potentially improve the nutrient status of
agricultural soils (Lal, 2004b), which in turn stimulates
SOC sequestration due to increased crop rhizodeposition
(Kuzyakov & Schneckenberger, 2004). In addition, it is
extremely important to consider C fluxes (CO2 and CH4)
because C outputs may offset the benefits of C inputs
from straw return. The balance of C inputs and losses
determines soil C stocks. In general, there is an upper
limit to the capacity for soils to store organic C, namely
SOC content can become saturated once C inputs and
outputs stay at equilibrium (Stewart et al., 2007). How-
ever, C saturation of soil mineral fraction is not well
understood (West & Six, 2007). Addressing C saturation
in soils still faces some problems, e.g., why/when does
the saturation of SOC occur? What are the fundamental
factors that are responsible for SOC saturation?
The effects of straw return on soil C dynamics
depend on many factors (West & Marland, 2002), of
which straw-C input rate is obviously the primary one.
Most cases have shown that there is a strong linear rela-
tionship between straw-C input rate and SOC seques-
tration (Duiker & Lal, 1999; Kong et al., 2005; Li et al.,
2010). Other factors, such as experimental duration
(Shen et al., 2007), N application rate (Jacinthe et al.,
2002), methods of disposing straw (Jarecki & Lal, 2003)
and soil conditions (e.g., soil microorganism, texture,
and aggregates) (Lugato et al., 2006; Bandyopadhyay
et al., 2010), are also important in explaining the large
differences in the magnitude and direction of SOC
changes after straw return. Hence, there is a critical
need to assess the effects of straw incorporation on soil
C cycle under various soil and management conditions,
which greatly benefits the development of proper straw
management strategies and contributes to SOC seques-
tration in global agricultural ecosystems. In particular,
SOC storage in paddy soils is significantly higher than
that in upland soils (Pan et al., 2003). However, the
responses of organic C pools to straw return in these
soils remain uncertain.
Many agricultural practices have been advocated to
mitigate global climate warming, but their positive ben-
efits to C sequestration may be largely offset by high
emissions of GHGs (Six et al., 2004). For example, N fer-
tilizer application generally enhances soil C sequestra-
tion, but it also increases N2O emission (Liu & Greaver,
2009; Lu et al., 2011b). When appraising agricultural
mitigating impact on global climate change in terms of
changes in C amount in agro-ecosystems, we need to
fully take into account of GHGs budgets and the net
GWP determined by GHGs emission and SOC seques-
tration (Lal, 2004b; Lu et al., 2009; Shang et al., 2011).
Although straw return is considered as one of the most
sustainable and economical C sequestration methods
(Triberti et al., 2008), to our knowledge, there are no
quantitative analyses integrating its positive or negative
effects on current global warming.
Meta-analysis offers a formal statistical method to
compare and integrate the results of multiple studies
and to draw general patterns at regional and global
scales (Gurevitch & Hedges, 1999; Luo et al., 2006). In
this study, we conduct a meta-analysis to address the
following five main questions/hypotheses: (i) to what
extent is soil C content altered by straw return? We
hypothesize that soil C sequestration may increase
under straw C input; (ii) what are the main factors that
influence the response of soil C to straw return? Specifi-
cally, how do initial C and clay content, aggregates,
and/or climate factors influence the results? We
hypothesize that lower initial C concentration may lead
to higher response of soil C; (iii) considering that soil
has a limited gain in C stocks, our hypothesis is soil C
saturation may occur under long-term straw C input;
(iv) can the increase in straw return stimulate GHGs
emission due to the increase in substrate quantity? and
(v) how do soil nutrient availability, crop production,
and agro-ecosystems C balance respond to straw C
input?
Materials and methods
Data sources
We collected data from 176 peer-reviewed experimental stud-
ies that reported changes in soil C dynamic induced by straw
return (Appendix S1). The selected publications were collected
by searching Web of Science (1900–2012, http://apps.webof
knowledge.com/) and China Knowledge Resource Integrated
Database (http://www.cnki.net/) with the keywords ‘straw
return or crop residue’ and ‘soil organic carbon or greenhouse
gases.’ Compiled database included the responses to straw
addition of 23 variables related to C pools, fluxes, and other
associated parameters (Appendix S2). To avoid publication
bias, three criteria were set to select the proper observations:
(i) amounts of straw return, land uses, and experiment dura-
tions were clearly stated; (ii) experimental sites of treatments
and controls had similar soil types and managements (e.g.,
cropping system, fertilization, and irrigation); and (iii) the
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
CARBON DYNAMIC AT STRAW RETURN 1367
means, standard deviations (or standard errors), and number
of replicates of chosen variables were directly reported or
could be calculated from reported data. Considering that
paddy rice and upland fields may respond differently to straw
return, we examined the effects of straw return on C dynamic
for these two types of land uses separately. In this study,
paddy fields/soils were referred to as arable land that was
submerged under water during one or more seasons of the
year for rice cultivation. Upland soils were all other irrigated
or nonirrigated agricultural systems where there was no regu-
lar submergence in whole year, except for temporary flooding
during irrigation or after heavy rain events. The effects of
straw incorporation (i.e., straw was ploughed/buried into the
soil) and cover (i.e., straw was mulched or covered on the sur-
face of soil) on SOC dynamics were tested in this analysis.
Experimental durations were classified into three categories:
short-term (1 ≤ duration ≤ 3 years), medium-term (3 < dura-
tion ≤15 years), and long-term (>15 years). Most compiled
data on CH4, CO2, and N2O fluxes reflected only short-term
effects (one or two growing seasons) in this analysis. For other
variables, studies with experimental durations less than 1 year
were excluded to avoid short-term noise.
Soil organic carbon concentration was converted from SOC
density with following equation (Pan et al., 2003):
SOCc ¼SOCp
H � q� 10ð1Þ
where SOCc is SOC concentration in g C kg�1 soil, SOCp is
SOC density in Mg C ha�1, q is soil bulk density in g cm�3,
and H is soil thickness in m. If the value of H was not clearly
pointed out, but soil-sampling depth was described by topsoil
or surface soil, we used 0.15 m to represent the value. In the
cases where data were missing, q was estimated following
Song et al. (2005)):
q ¼ 1:3770� e�0:0048�SOCc : ð2Þ
Data reported as soil organic matter concentrations were
converted to SOC concentration by multiplying with a conver-
sion factor of 0.58 (Pan et al., 2003). Given C concentration of
crop straw not available, straw-C input was estimated by
assuming that straw contained 40% C on average.
Data analysis
Meta-analysis was used to determine changes in soil C cycle
after straw return to soils in various cropping systems
(Hedges et al., 1999; Luo et al., 2006). Twenty-three variables
related to C pools, fluxes, important soil physicochemical
properties, and crop production were selected. Means (M),
standard deviations (SD), and sample sizes (n) of selected vari-
ables were extracted for each case. If only the standard errors
(SE) were given, SD was calculated by:
SD ¼ SEffiffiffin
p: ð3Þ
The natural log-transformed response ratio (lnRR) was
employed to reflect the effects of straw return on soil C pools,
gas fluxes, and soil properties in agricultural systems, and it
was calculated by following equation (Hedges et al., 1999):
lnRR ¼ lnðXt=XcÞ ¼ lnðXtÞ � lnðXcÞ ð4Þwhere Xt and Xc are means of the treatment and control
groups for variable X, respectively, and the variance (v) of X
is computed as:
v ¼ SD2t
ntX2t
þ SD2c
ncX2c
: ð5Þ
where SDc and SDt are standard deviation of control and treat-
ment groups, and nc and nt are sample sizes of control and
treatment, respectively.
In addition, the weighting factor (wij), weighted response
ratio (RR++), the standard error of RR++ (S), and 95% confi-
dence interval (CI) of RR++ were calculated as below (Curtis &
Wang, 1998; Luo et al., 2006):
wij ¼ 1
vð6Þ
RRþþ ¼
Pmi¼1
Pkij¼1
wijRRij
Pmi¼1
Pkij¼1
wij
ð7Þ
SðRRþþÞ ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
Pmi¼1
Pkij¼1
wij
vuuuutð8Þ
95%CI ¼ RRþþ � 1:96SðRRþþÞ: ð9ÞIf the 95% CI of RR++ for a given variable overlapped with
zero, the response of that variable to straw return represented
no significant differences between straw treatment and control
(Luo et al., 2006). The percentage change of a variable was cal-
culated by the equation of eRRþþ � 1� �� 100%.
Frequency distributions of lnRR were plotted to reflect the
variability of straw return effects among different studies by a
Gaussian function (i.e., normal distribution) (Luo et al., 2006):
y ¼ aeðx�lÞ22r2 ð10Þ
where y is the frequency of lnRR values within an interval, x
is the mean of lnRR for that interval, l and r2 are the mean
and variance of all lnRR values, respectively, and a is a coeffi-
cient indicating the expected number of lnRR at x = l.For assessing C budget caused by straw return, we esti-
mated responses of SOC, GHGs and DOC per unit straw-C
added by an equation adapted from Liu & Greaver (2009) and
Xie et al. (2010).
FX ¼ 1
n
Xni¼1
ðXti � XciÞCIRi
: ð11Þ
where FX represents contribution ratios of variable X (SOC,
GHGs or DOC) (Mg C or kg N ha�1 yr�1 per Mg
C ha�1 yr�1), Xti and Xci are values of variable X in the ith
sample of treatment and control, respectively, n is the sample
size, and CIR is straw-C input rate (Mg C ha�1 yr�1).
The net global warming potential (GWP) balance caused by
straw return, presented as CO2-C equivalent (CO2-Ceq), was
calculated based on the GWP of different GHGs as below:
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
1368 C. LIU et al.
netGWP ¼ �Fsoc � Fresidual�c þ FCH4� 16
44� 25þ FN2O
� 12
28� 1000� 298 ð12Þ
Where FSOC, FCO2, FCH
4, and FN
2O are contribution ratios of
SOC, CO2, CH4, and N2O emission. Fresidual-C is the mean
straw residual-C in soil and is set to 20% of originally incorpo-
rated straw-C following Jenkinson (1971) and Shields & Paul
(1973). The values of the GWPs of CH4 and N2O relative to
CO2 are 25 and 298 at a 100-year time horizon, respectively
(IPCC, 2007).
Sigma Plot 11.0 was used to fit data to normal distribution.
Regression analysis was applied to examine relationships of
lnRR of SOC concentration with lnRRs of other variables (soil
active C fraction, grain yield, and macro-aggregates) and
experimental factors (straw-C input rate, experimental dura-
tion, N fertilization rate, and soil depth). One-way ANOVA was
employed to examine whether the RR++ of variables differed
significantly between upland and paddy soils, straw disposal
methods of SOC (incorporation vs. cover), or among soil tex-
tures (clay, sand, and silt). We also calculated the total hetero-
geneity in lnRR among studies (QT); and in the case of
comparisons among groups, the differences in cumulative
effect size (QM) and residual error (QE) among groups (Rosen-
berg et al., 2000). All statistical analyses were performed using
SPSS 16.0.
Results
Responses of soil organic carbon to straw return
Straw-C input into the soil-enhanced SOC concentra-
tion of bulk soil and active SOC fractions (Fig. 1). The
RR++ of SOC across all the 343 pairs of comparisons
was 0.120 � 0.007 (RR++ � 95% CI, same as below;
Fig. 1a), meaning that straw return led to a 12.8%
increase in SOC concentration compared with the con-
trol. The lnRR of SOC exhibited a great variability
among different studies, with a range from �0.467 to
0.529. The frequency distribution of lnRRs could be
characterized by a Gaussian normal distribution
(R2 = 0.916 and P < 0.001). The RR++ (0.120) of meta-
analysis approach was consistent with that from the l
(a) (d)
(b) (e)
(c) (f)
Fig. 1 Frequency distributions of response ratios (lnRR) for soil organic C [SOC, (a)], microbial biomass C [MBC, (b)], particulate
organic C [(POC, (c)], dissolved organic C [(DOC, (d)], easily oxidizable C [(EOC, (e)], and light fraction organic C [(LFOC, (f)]
responses to straw-C input. The solid curve is a Gaussian distribution fitted to frequency data. The vertical dashed line is at lnRR = 0.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
CARBON DYNAMIC AT STRAW RETURN 1369
value (0.115) simulated by a Gaussian function model
(Fig. 1a, Appendix S2). The response was inconsistent
among studies, as indicated by a significant QT value
(QT = 1377.12 and P < 0.05; Appendix S3). The effects
of agricultural land use (paddy rice vs. upland) on the
accumulation of SOC under straw return were not sig-
nificant, with a RR++ of 0.117 � 0.010 in paddy soils
and 0.125 � 0.010 in upland soils. We also found that
clay loam soil generally had a lower effective size
(0.100 � 0.018) than that of sandy loam (0.157 � 0.016)
and silt soil (0.151 � 0.016). Meanwhile, RR++ values of
microbial biomass C (MBC), easily oxidizable C (EOC)
and dissolved organic C (DOC) were 0.267 � 0.011,
0.242 � 0.027 and 0.244 � 0.021, respectively, which
were all significantly different from zero (P < 0.001;
Fig. 1b, d, and e). The percentage increases were
similar in light fraction organic C (LFOC) and particu-
late organic C (POC), but higher than in other C fraction
(Fig. 1c and f). The increments of MBC, DOC, LFOC,
and POC in upland soils were larger than those in
paddy soils, while the increment of EOC in upland soils
was smaller than that in paddy soils (P > 0.05; Fig. 2).
Responses of CH4, CO2, and N2O to straw return
Responses of CH4, CO2, and N2O to straw return were
all significantly heterogeneous among studies (Appen-
dix S3). In comparison with the control group, straw
return significantly increased C fluxes, with a RR++ of
0.745 � 0.023 for CH4 and 0.328 � 0.027 for CO2 in
agricultural soils, both were significantly greater than
zero (P < 0.001; Fig. 3a and b). For soil CH4, CO2, and
Fig. 2 Weighted response ratios (RR++) of variables in agricultural system in response to straw return. Bars represent RR++ � 95% CI.
The sample size for each variable is shown next to the bar.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
1370 C. LIU et al.
N2O, the frequency distribution of lnRRs could be char-
acterized by a Gaussian normal distribution (all
R2 > 0.900 and P < 0.001; Appendix S2). For the three
GHGs, the RR++ of meta-analysis was about consistent
with those from the l values from fitted Gaussian equa-
tions (Fig. 3, Appendix S2). An increase in CO2 fluxes
in paddy soils (0.412 � 0.039) was significantly higher
than that in upland soils (0.245 � 0.038) (P = 0.026;
Appendix S3). In this analysis, data were not sufficient
to estimate CH4 emission for upland cropping systems.
By contrast, N2O emission significantly decreased with
straw return by a RR++ of 0.123 � 0.020 (P < 0.001;
Fig. 3c). Fig. 2 also showed that N2O emission in the
two types of agricultural systems responded differently
to straw return. A significant increase in N2O
(RR++ � 95% CI = 0.079 � 0.048, P < 0.01) was found
in upland soils, while a drop was observed in paddy
soils (RR++ � 95% CI = �0.164 � 0.022, P < 0.001).
Responses of crop yield and soil properties to straw return
Crop yield greatly increased by straw return in both
upland and rice paddies (P < 0.001; Fig. 2), with a RR++
of 0.180 � 0.021 for upland crops and 0.084 � 0.015 for
paddy rice. Global average RR++ of crop yields was
0.116 � 0.012.
Overall, straw return improved soil nutrient levels.
The RR++ values of these soil nutrients were in the
range of 0.033 to 0.122, all being significantly greater
than zero (Fig. 4a–f). The increase in total N (TN) and
available N (AN) were similar, being higher in straw-
return treatment than in control. The total P (TP) and
available P (AP) were both increased due to straw
return. Increased percentage was lower in total K (TK)
than in available K (AK). Agricultural land use did not
significantly affect the changes in TN, TK, AN, and AK.
The changes in TP, however, showed significant differ-
ences between upland and paddy soils, with a RR++ of
0.088 � 0.020 in upland soils and �0.018 � 0.042 in
paddy fields (Fig. 2).
Straw addition slightly decreased soil bulk density,
with a RR++ of �0.020 � 0.012 (P < 0.01; Fig. 5a). Soil
porosity was increased due to straw input and the cor-
responding RR++ value was 0.079 � 0.036 (P < 0.001;
Fig. 5c). However, soil pH exhibited only a minor
change (RR++ � 95%CI = �0.003 � 0.020, P > 0.05;
Fig. 5b) in response to straw return. Overall, agricul-
tural land use has no significant impacts on soil bulk
density, porosity, and pH.
Straw input affected the distribution of soil aggre-
gates (Fig. 5d and e). Compared with those in controls,
soil macro-aggregates, and mean weight diameter
(MWD) increased in straw treatment, while soil micro-
aggregates decreased (all P < 0.001). The changes in
soil macro-aggregates, micro-aggregates, and WMD
showed no significant differences between upland and
paddy systems.
Factors affecting the responses of soil organic carbon tostraw return
Figure. 6 showed that straw-induced changes in SOC
significantly decreased with soil depth (b), and clay
content (d), while significantly increased with macro-
aggregates (h). Our results indicated a significantly
negative linear relationship between lnRR of SOC and
initial SOC concentration under straw addition
(R2 = 0.062, P = 0.001; Fig. 6c), whereas a significantly
positive linear for straw-C input rate (R2 = 0.025,
P = 0.003; Fig. 6a). The lnRR of SOC was also signifi-
cantly related to crop yields (Fig. 6g). A significantly
linear regression between the lnRR of soil MBC and the
lnRR of total SOC was shown in Fig. 6f, which was an
(a)
(b)
(c)
Fig. 3 Frequency distributions of response ratios (lnRR) for
CH4 (a), CO2, (b) and N2O (c) responses to straw return. The
solid curve is a Gaussian distribution fitted to frequency data.
The vertical dashed line is at lnRR = 0.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
CARBON DYNAMIC AT STRAW RETURN 1371
example for representing the relationship between
active SOC fractions and total SOC (P < 0.001). In addi-
tion, N fertilization rate, mean annual temperature
(MAT), and mean annual precipitation (MAP) were not
significantly linear with SOC changes (Appendix S4).
Soil organic carbon accumulation greatly differed
between methods of straw disposal (incorporation vs.
cover). The RR++ of SOC for straw incorporation
(0.123 � 0.009) was slightly higher than that for straw
cover (0.110 � 0.013; Appendix S3). In short-term or
long-term experiments, straw-induced changes in SOC
had no significant change with experimental duration
(Fig. 6e). Within medium-term experiments, however,
the RR++ of SOC significantly increased with the dura-
tion after straw return (P < 0.05).
Relationships between straw C input, duration, and SOCsequestered
The relationship between annually sequestered SOC (y:
Mg C ha�1 yr�1) and duration (x: year) was described
by the following equation: y = 4.64/x�0.41 (R2 = 0.372,
P < 0.0001; Fig. 7a). Soil C saturated at about 12 years
when C loss was equal to C input. Significant positive
regression relationship was found between cumulative
straw C input and SOC sequestered (R2 = 0.08,
P < 0.0001; Fig. 7b), which could be represented by the
fitted equation: y = 11.0 9 (1�e(�0.013x)).
Changes in agro-systems C balance at global scale
Table 1 showed that the response of SOC sequestration
to straw return was similar between in paddy
(0.160 � 0.034 Mg C ha�1 yr�1 per Mg C ha�1 yr�1)
and upland soils (0.166 � 0.024 Mg C ha�1 yr�1 per Mg
C ha�1 yr�1). Fluxes of CO2 increased by 0.518 � 0.101
and 0.409 � 0.068 Mg C ha�1 yr�1 per Mg C ha�1 yr�1
in paddy and upland soils, respectively. Methane (CH4)
was enhanced by 0.124 � 0.025 Mg C ha�1 yr�1 per Mg
C ha�1 yr�1 in paddy soils. N2O emission showed a
positive response to straw addition in upland soils, but
a negative response in rice paddies.
(a) (d)
(b) (e)
(c) (f)
Fig. 4 Frequency distributions of response ratios (lnRR) for Total N [(TN, (a)], Total P [(TP, (b)], Total K [(TK(c)], Available N [(AN,
(d)], Available P [(AP, (e)], and Available K [(AK, (f)] responses to straw addition. The solid curve is a Gaussian distribution fitted to
frequency data. The vertical dashed line is at lnRR = 0.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
1372 C. LIU et al.
At global scale, 0.124 Pg C was sequestered annually
in world agricultural soils by straw return (Table 1).
Changes in net GWP due to straw return was estimated
to be �0.363 Mg C ha�1 yr�1 per Mg C ha�1 yr�1 in
upland soils, and 0.761 Mg C ha�1 yr�1 per Mg
C ha�1 yr�1 in rice paddies, meaning that a total of
0.137 Pg C yr�1 was increased by straw addition in
agricultural soils of the world.
Discussion
The effects of straw return on SOC sequestration
Soil organic carbon sequestration in cropping systems
has been considered as a cost-effective and environmen-
tally friendly strategy for sequestrating anthropogenic
CO2 emission (Lal, 2004a). Although numerous studies
have demonstrated that SOC sequestration was influ-
enced by extra organic C input, the magnitude and even
the direction of this response varied among studies.
Overall, we found that straw return significantly
increased SOC concentration by 12.8%, compared with
5.1% increase in SOC after adoption of no tillage
(Angers & Eriksen-Hamel, 2008), and to 3.5% increase in
SOC due to N fertilization (Lu et al., 2011b), suggesting
that straw-C input might further promote C accumula-
tion in arable soils. This could be explained by increased
crop-derived C input into the soil, as supported by the
positive linear-related between the lnRR of SOC and
straw-C input rate (Fig. 6a). Similar findings have been
reported from other field experiments at site scale (Dui-
ker & Lal, 1999; Kong et al., 2005; Li et al., 2010). Our
results also showed that crop yield was significantly
and positively linear with organic C input. High crop
productivity meant greater organic C input into the soil,
which consequently lead to additional C sequestration
in agricultural systems (Jarecki & Lal, 2003; Cai & Qin,
2006). Thus, our result supported our first hypothesis
and confirmed that application of straw was an effective
measure to enhance SOC level in croplands.
Straw return-induced SOC increase appeared to be
higher in upland than that in paddy soils, though the
(a) (d)
(b) (e)
(c) (f)
Fig. 5 Frequency distributions of response ratios (lnRR) for soil bulk density (a), pH (b), porosity (c), macro-aggregate (d), micro-
aggregate (e) and mean weight diameter (MWD, (f) responses to straw-C input. The solid curve is a Gaussian distribution fitted to
frequency data. The vertical dashed line is at lnR = 0.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
CARBON DYNAMIC AT STRAW RETURN 1373
difference was not statistically significant. This result
was not consistent with current knowledge that paddy
soils may have generally higher potential of SOC
sequestration than upland soils (Pan et al., 2003; Song
et al., 2005). In general, paddy soils have higher SOC
background (namely higher initial SOC concentration)
already than upland soils, and therefore, the same
amount of returned straw C would cause a less percent-
age change in SOC of paddy fields. Because the anaero-
bic condition during rice growing season in paddy soils
restricts SOC or straw decomposition, SOC in paddy
soils is more readily decomposable than that in upland
soils (Cui et al., 2012), leading to a significantly higher
SOC decomposition in paddy soils in consequent non
rice growing season. Interactions among various pro-
cesses or mechanisms are responsible for the complex
responses of paddy and upland soils to straw return.
Additionally, the effect of straw return on SOC
sequestration was influenced by the way how the
return was conducted. For instance, we found that
incorporating straws into soils tended to increase SOC
more evidently than covering straws on soil surface.
Although straw incorporation by tillage may strongly
disturb the soil and cause some losses of SOC due to
exposition to oxygen (Angers & Eriksen-Hamel, 2008;
Luo et al., 2010), the increase in contacting between
straw and soils can stimulate more straw C to trans-
form into organic C in the soil.
Our analysis suggested that changes in all active SOC
fractions, including MBC, POC, DOC, EOC, and LFOC,
were significantly associated with changes in SOC.
Haynes (2000) also reported that MBC and LFOC
increased linearly with SOC. However, the rates of
increases in active soil C fractions were higher than that
in total SOC, indicating that the active components of
SOC responded more sensitively to straw return.
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
Fig. 6 Relationships between the response ratio (lnRR) for the response to straw return of soil organic C (SOC) level and straw-C input
rate (a), soil depth (b), initial SOC concentration (c), clay content (d), experimental duration (e), lnRR of MBC (f), lnRR of yield (g) and
lnRR of macro-aggregates (h).
(a) (b)
Fig. 7 The relationship between straw C input, duration, and SOC sequestered in agricultural soils. (a) annual SOC sequestered vs.
duration (b) SOC sequestered vs. cumulative straw C input.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
1374 C. LIU et al.
Similarly, Malhi et al. (2011) noticed that the change in
soil LFOC due to straw management was more remark-
able than total SOC. The reason was possibly that crop
residues mainly provided readily decomposable sub-
strates to the soil (Chen et al., 2009; Schulz et al., 2011).
All these findings suggested that active SOC fractions
could be proper indicators of the direction and magni-
tude of soil C cycling after straw return.
Macro-aggregates played an important role in the
accumulation of SOC (Six et al., 2002; Kong et al., 2005).
It was expectable that returning crop residues into the
soil could greatly affect the formation of soil macro-
aggregates (Duiker & Lal, 1999; Benbi & Senapati,
2010). Our synthesis revealed that straw addition
significantly increased the amounts of soil macro-aggre-
gates. There may be at least two explanations for the
straw-induced changes. First, straw decomposition
directly increased particles or colloids that were associ-
ated with mineral matter binding micro-aggregates into
macro-aggregates. Second, increased microbial biomass
(Fig. 1b) and activity (Pan et al., 2009) due to straw
addition might have led to a greater production of
microbial-derived binding agents and consequently
had a greater macro-aggregate formation rate. In addi-
tion, most of accumulated C occurred in mineral-associ-
ated fraction of macro-aggregates (Jastrow, 1996),
which slowed down SOC decomposition and promote
SOC stabilization (Bandyopadhyay et al., 2010).
Clay content has been considered as a crucial factor
to influence the capacity of soil to store C (Six et al.,
2002). Our data set covered a wide range of clay content
from 3% to 65%, and the result indicated that straw-
induced changes in SOC had a negative relationship
with clay content. A similar previous result was
observed by Huang et al. (2002), who reported that a
strong negative relationship existed between the
decomposition of wheat straw and clay content. One
possible reason was that biodegradation rates of straws
were probably slower in the soil with higher clay con-
tent (Balesdent et al., 2000), which may lead to lower
efficiency of C transferring from straws to soil minerals,
i.e., less humified SOC formed in these soils.
Our results showed that MAT and MAP were not
significantly linear-related with SOC changes in agri-
cultural soils (Appendix S4). Luo et al. (2010) also
pointed out that the pattern of change in soil C after the
adoption of no-tillage might not be related to climate
(i.e., MAT and MAP). These suggested that the agricul-
tural system was an intensively human-managed one,
and climate conditions only slightly influenced C
dynamics in agricultural soils.
May soil C saturation occur under straw return?
The effects of straw return were not evident in short-
term field experiments (1–3 years). In medium-term
Table 1 Estimates of net changes in the C balance of global agro-systems caused by straw return
Straw-induced changes per unit straw-C added
(Mean � SE; Mg C or kg N ha�1 yr�1 per Mg C ha�1 yr�1)
FSOC FCO2
FCH4
FDOC FN2O
Straw†
residual Sum‡
Upland 0.166 � 0.024 0.518 � 0.101 N/A 0.008 � 0.002 0.024 � 0.018 0.200 0.889
Paddy soils 0.160 � 0.034 0.409 � 0.068 0.124 � 0.025 0.0005 � 0.0003 �0.044 � 0.055 0.200 0.894
net GWP balance§ (CO2-Ceq) attributed to straw-C added (Mg C ha�1 yr�1 per Mg C ha�1 yr�1)
Upland �0.363
Paddy soils 0.761
Global scale
balance
under straw return
(Pg C yr�1)
net SOC†† net GWP
Agriculture
systems
�0.124 �0.137
†The percentage of straw-C residual was according to Jenkinson (1971) and Shields & Paul (1973).
‡Sum = FSOC + FCO2+ FCH
4+ FStraw residual + FDOC.
§net GWP balance (CO2-Ceq) attributed to straw-C added was estimated by using Eqn (12) (detail was provided in the Methods
section). The value of net SOC and GWP represent a net C source (positive) or sink (negative).
††net SOC = �ΣFSOC 9 S 9 CIRT. S is the area of arable land, and world areas of upland and paddy soils are 12.2 9 108 ha and
1.6 9 108 ha in 2009, respectively (data from http://www.fao.org). CIRT is the mean straw-C input rate (Mg C ha�1 yr�1) in the
world. If 50% of crop residue was returned to soil, the CIRT was estimated to be 0.517 and 0.755 (Mg C ha�1 yr�1) for upland and
paddy systems, respectively (Lal, 2008).
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
CARBON DYNAMIC AT STRAW RETURN 1375
experiments (3–15 years), SOC contents significantly
increased with the duration of straw return. In long-
term experiments (>15 years), however, no obvious
changes in lnRR with time could be seen, indicating a
possible saturation of SOC levels after long-term appli-
cations of straw. These results could be envisaged as
the evidence of C saturation. Three possible mecha-
nisms might underpin organic C saturation in the soil:
(1) soil capacity to maintain organic C, (2) sequestration
duration, and (3) straw C input rate. The capacity of
soil to capture C mainly depend on clay content (i.e.,
chemical stabilization), aggregates (i.e., physical protec-
tion), and recalcitrant compounds in SOM (i.e., bio-
chemical stabilization; Six et al., 2002). Paddy soils
generally contain more recalcitrant compounds of SOM
than upland soils (Zhou et al., 2009), which means that
straw C may have more opportunity to transform resis-
tant components of SOC in rice paddies. Pre-experi-
mental SOC concentrations can also influence the
ability of straw inputs to result in soil C saturation
achieved. Our result indicated a significantly negative
linear-related between the response ratio of SOC and
initial C concentration under straw addition (Fig. 6c),
suggesting the influence of initial SOC concentration on
the effects of straw return (supported our second
hypothesis). A soil with a low initial soil C content has
a large saturation deficit that lead to a faster initial C
sequestration rate and a longer total duration to reach
the same C saturation (West & Six, 2007). Duration of C
sequestration has a significant impact on soil C satura-
tion, as longer duration favoring straw C to transform
into SOC. Our study revealed that 12 years on average
were required for the soil to reach C saturation (or a
steady state of C) under continuous straw return. A
similar study showed that with adoption of land restor-
ative measures, usually the rate of C sequestration
would be small beyond 15 year (Akala & Lal, 2000).
Whereas West & Six (2007) estimated that soil C satura-
tion might occur over a period of 26 years under inten-
sive rotation and 21 years under tillage cessation in
agronomic systems, both longer than that under straw
return in this study (12 year). High variability was seen
in the response of SOC to straw return at early stage,
and this variability declined along with time (Fig. 7a).
The change percentage of SOC was commonly small at
the early stage of straw return. It is still a technical chal-
lenge to detect a small change in SOC. A large number
of samples were required to obtain a reliable estimation
of SOC change (Conen et al., 2003). For most reported
long-term experiments, heavy sampling was practically
impossible. Other two possible reasons for the high var-
iability of SOC response at early stage of straw return
are disturbances of experimental implementation and
the quick release of unstable SOC pools. Cumulated
quantity of straw C input generally resulted in
increased supply of organic material and allowed SOC
to accumulate (Fig. 7b). However, changes in straw C
input rate may promote or retard soil C saturation but
do not alter soil C capacity. Luo and Weng (2011) pro-
posed that the terrestrial C cycle might present as a
dynamic disequilibrium, namely a disturbance caused
temporal changes in C source or sink at yearly and dec-
adal scales but had no impact on long-term C seques-
tration unless the disturbance regime was altered.
According to their framework, our results indicated
that addition of plant C could only result in a limited
gain in soil C stocks in a long term (Stewart et al., 2007;
Lu et al., 2009).
The response of GHGs fluxes to straw C input
Among GHGs, CO2 is one of the most significant con-
tributors to anthropogenic climate change. CO2 produc-
tion in arable soil could be mainly attributed to crop
residue decomposition, roots respiration, and minerali-
zation of SOC (Luo & Zhou, 2006). Our analysis indi-
cated that straw return had significant increased CO2
effluxes in both upland and paddy soils. This should be
partially due to increased plant productivity after straw
return, which often resulted in great root respiration
(Iqbal et al., 2009). Also, the increase in microbial bio-
mass under straw return may have positive conse-
quences in C fluxes. Lou et al. (2004) reported that
enhanced soil MBC could stimulate the decomposition
of SOC and straw, thereby stimulate C emission. Addi-
tionally, increases in soil moisture due to straw return
might accelerate CO2 emission, as suggested by Jabro
et al. (2008). Enhancement of soil porosity could also
improve soil diffusivity and contribute to CO2 emission
from the soil surface. Straw return induced increase in
soil moisture and soil porosity are both possible, but in
different circumstances. In sandy soils, straw return
induced increase in SOC could possibly increase water-
holding capacity of the soil due to the high water
absorption of organic matter and soil aggregation from
the banding function of SOC. In compact soil where
clay content is high, straw return induced increase in
SOC necessarily lead to an increase in soil porosity.
It was notable that CH4 emission in paddy soils was
increased by 110.7% under straw return (Fig. 3), which
was consistent with previous integrated analysis
reported by Xie et al. (2010). A number of studies have
demonstrated that CH4 emission was largely influenced
by the input of organic substrates under anaerobic con-
ditions (Cochran et al., 1997; Naser et al., 2007; Ma et al.,
2008). Addition of straw could selectively enhance the
growth of particular methanogenic populations and
therefore stimulated rates of CH4 production (Conrad
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
1376 C. LIU et al.
& Klose, 2006). In addition, anaerobic decomposition of
straw does not only provide substantial methanogenic
substrates, but also speed up a decrease in soil Eh,
favoring CH4 production (Cai et al., 1997; Ma et al.,
2008). Therefore, regulating soil moisture is an effective
means to reduce CH4 emission in paddy soils. Middle
season drainage of rice paddies often led to significant
drops in seasonal CH4 fluxes and has been recommend
as an important irrigation measure for mitigating global
warming (Zou et al., 2005; Xie et al., 2010).
Soil N2O produced naturally through microbial pro-
cesses of nitrification and/or denitrification, which
were profoundly influenced by soil N availability (e.g.,
NH4+), soil water content/oxygen and organic C sup-
ply (Chapin et al., 2002; Khalil et al., 2004). Our results
showed that straw return in upland soils increased
N2O emission, while a remarkable decline was seen in
rice paddies. In upland soils, straw return can provide
sufficient substrate for soil microbes and result in
enhanced soil N availability and hence an acceleration
of microbial processes. The increases in N2O flux might
be attributed to stimulated nitrification and denitrifica-
tion (Otte et al., 1996). A possible explanation, during
rice growth, is that soil water layer and low oxygen
could restrain nitrification and trap N2O in rice paddies
and thus inhibit its release to atmosphere (Cai et al.,
1997).
It should be noted that compared with CO2, the
responses of CH4 to straw return show largest variabil-
ity (Fig. 3). The lower concentration of trace gases and
considerably large difference in CH4 flux from different
sites (Appendix S2) may cause this variability.
Responses of soil quality and crop productivity to strawreturn
Understanding the relationships between agronomic
productivity, soil quality, soil organic matter, and
agronomic productivity is very important for the sus-
tainability of agricultural systems (Bhogal et al., 2009;
Sommer et al., 2011). The advantages of using straw to
improve soil quality and agricultural productivity
have been increasingly recognized (Jarecki & Lal,
2003). Our meta-analysis showed that crop yield could
increase by 12.3% after straw return, suggesting the
benefits of straw return to cropland productivity.
Crop straw is considered as a resource containing
readily available nutrients required for crop growth
(Pathak et al., 2006). Given that plant growth is usu-
ally limited by soil nutrients availability (Vitousek &
Howarth, 1991; Lu et al., 2011a), our results clearly
indicated that increased soil nutrient (particularly
available nutrients N, P, K) pools under straw return
may favor crop growth.
Straw return resulted in evident improvements in
soil physical conditions, as indicated by changes in soil
bulk density, total porosity, and water-holding capac-
ity, possibly due to enhanced SOC levels (Duiker & Lal,
1999; Zhu et al., 2010). Straw addition also improved
soil biological functions, which was supported by
increased microbial biomass (Bhogal et al., 2009). Over-
all, our synthesis suggested that straw return was an
effective measure to enhance agricultural soil quality
and crop productivity, which was especially important
for the sustainable development of agriculture.
Carbon balance in agro-ecosystems. Our analysis indi-
cated that straw return has promoted SOC accumula-
tion in agricultural soils. On average, 16.3% of organic
C in returned straws has been transferred into SOC,
which was slightly lower than reported transformed
ratio of 18.8% in China’s agricultural soils (Lu et al.,
2009). At global scale, the annual SOC sequestration by
straw return in agricultural ecosystems could be 0.124
Pg C, about 1.4% of global fossil-fuel emissions in 2008
(8.7 Pg C yr�1, Le Qu�er�e et al., 2009). This was close to
estimated C sequestration potential by adoption of con-
servation tillage and crop residue management (0.150–0.175 Pg C yr�1, Lal & Bruce, 1999).
After a certain time of straw return, a part of straws
will still exist in the soil in the form of plant residues.
Jenkinson (1971) and Shields & Paul (1973) reported
that after 4–5 years of decomposition in the field, the
residual straw was still 20% of the original straws
returned and subsequent decomposition of this residue
was very slow. In common practice of SOC determina-
tion, this part of C was not included. The potentially
significant contribution of this residual straw to the mit-
igation of global warming has been commonly ignored
and consequently GHGs mitigation capability by straw
return might be underestimated. Therefore, the overall
estimation of straw return in agricultural systems
should include the residual C in the estimates of global
net change. Our estimations indicated a net source of
GHGs at global scale in rice paddies, but a net sink in
upland systems. Increased CH4 emission largely offset
the C sink in paddy system. Overall, the global net
CO2-C mitigation capacity in agricultural soils was esti-
mated to be 0.137 Pg C yr�1. This indicated that
although CH4 and N2O emissions were taken into
account, straw return into agricultural soils could
reduce global climate warming. To achieve more bene-
fit of C sequestrating by straw return, agricultural
systems should be properly managed to reduce
straw-induced GHGs emission (especially CH4 emis-
sion in rice paddies). Two means have been proposed
to avoid high CH4 emission triggered by straw amend-
ment. One was to alter straw amendment from rice
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
CARBON DYNAMIC AT STRAW RETURN 1377
growing season to non-rice growing season (Xie et al.,
2010; Shang et al., 2011). Another one was to regulate
water regime (e.g., midseason drainage and intermit-
tent irrigation) during rice growing season (Zhang
et al., 2011) to maximize the oxidation/minimize the
production of CH4. Midseason drainage could reduce
the total CH4 emission by ca. 50% (Yagi et al., 1996; Zou
et al., 2005). If this reduction could be achieved in all
paddy soils of the world, the global net sink of agricul-
tural soils was up to 0.205 Pg C per year.
Limitation of the study and improvement needed. Although
the meta-analysis provides a statistical approach to cal-
culate weighted response ratios across experiments
and sites, the overall effects of straw return on soil C
dynamics may be synthesized with uncertainties due
to the inherent limitations of methodologies used in
such analyses and/or of data quality deficiencies
among different studies. First, published data used
here may bias toward those studies with strong straw
return effects, because such studies might have been
more frequently reported (Gurevitch & Hedges, 1999).
According to soil map of the world (FAO, 2007) and
global distribution of straw return experiments
(Appendix S5), compiled database extracted in our
meta-analysis might have mostly come from multiyear
studies conducted on medium-fertility soils, while
those on fertile or infertile soils might have been
ignored because of insignificant results. The lack of
studies from infertile and fertile soils may have
influenced our conclusions. Second, the high heteroge-
neity of data sets from different studies, e.g., different
sampling depths, experimental durations and methods
of straw disposal, also influenced the overall effects of
straw return on soil C dynamics. GHGs fluxes usually
showed high temporal or spatial variability, especially
in highly disturbed agro-ecosystems. However, sample
sizes and experimental durations were often not suffi-
cient to represent spatial/temporal heterogeneity.
Also, most studies reporting CH4, CO2, and N2O emis-
sions lasted only one or two growing seasons. In this
work, the average annual flux was calculated from the
data of growing season. However, seasonal values
may not actually present the annuals (Johnson et al.,
2000). The ignorance of changes in nongrowing season
could underestimate/overestimate GHGs fluxes and
GWP because the responses of annual or growing sea-
son CO2, CH4, and N2O emission may be not consis-
tent. In addition, data of soil bulk density were not
available in some experimental sites. In this study,
bulk density was estimated by the relationship
between bulk density and SOC concentration using
Eqn (2), which might not actually represent true val-
ues. Uncertainties in estimating SOC stocks also might
influence the CO2-Ceq estimates. Nevertheless, together
with the modeling of Gaussian distribution and the
regression analysis, the method of meta-analysis
offered a powerful statistical analysis, and these
uncertainties and limitations might be unlikely to
change the response patterns of soil C dynamics in
Fig. 8 A simple mechanism for the responses of soil C dynamics to straw return in agro-ecosystems.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
1378 C. LIU et al.
agricultural systems to straw return. Moreover,
according to our results, agricultural land uses, initial
soil C concentration, experimental duration all influ-
enced the overall pattern of soil C dynamics in rela-
tion to straw return. Thus, in future cropland
management, differentiated straw return strategies are
needed in upland and paddy rice soils to achieve the
goal of mitigating GHGs emission. The different
responses of land uses (upland vs. paddy soils) to
straw return need to be incorporated into agro-ecosys-
tem models to simulate soil C dynamics. In addition,
although our meta-analysis pointed out the potential
mechanism for how straw C input influence C dynam-
ics in croplands, given that the agricultural system
was an intensively human-managed system, further
studies focusing on interaction of straw return with
other cropland management would be highly needed
to reveal how long-term straw C input, tillage, fertil-
ization and irrigation interact to regulate cropland C
dynamics.
Conclusions
In summary, our results showed that straw return sig-
nificantly enhanced SOC accumulation in agro-ecosys-
tems. Straw-C input rate and soil clay content were
recognized as two major factors affecting the effects of
straw return on SOC. The relationship between
annual SOC sequestered and duration suggested that
soil C saturation would occur after 12 years of straw
C input. Straw return also improved overall soil qual-
ity, nutrient availability and crop yields. All these
changes suggested a net benefit of straw return to C
sequestration and crop productivity of agricultural
soils. However, it should be noted that straw return
also stimulated CH4 and CO2 emissions, which
deserves more attention by land managers in the con-
text of global warming. All these changes support a
conceptual framework in which there are positive
feedbacks between straw return, C cycles and soil
quality in agro-ecosystems (Fig. 8). Specifically, effec-
tive management strategies should be adopted to
reduce straw-induced GHGs emissions to realize
GHG mitigation potential of straw return in rice pad-
dies, e.g., optimized water regime and improved N
efficiency. Given that the total GWP increased due to
CO2, CH4, and N2O emissions, agricultural soils
amended with crop straws may play an important
role in mitigating future global climate warming. The
responses of SOC and GHGs emissions to straw
return as revealed by this synthesis can be potentially
useful for future cropland surface models for better
understanding and predicting the agro-ecosystem C
cycle feedbacks to straw C input.
Acknowledgements
We are grateful to the anonymous reviewers for their insightfulcomments, which greatly improved the manuscript. We thankthe authors whose data and work were included in this meta-analysis. This research was financially supported by the Minis-try of Science and Technology of China (Grant No.2010CB950604), and China National Natural Science Foundation(Grant No. 31070461; 31100352).
References
Akala V, Lal R (2000) Potential of mine land reclamation for soil organic carbon
sequestration in Ohio. Land Degradation and Development, 11, 289–297.
Angers D, Eriksen-Hamel N (2008) Full-inversion tillage and organic carbon distribu-
tion in soil profiles: a meta-analysis. Soil Science Society of America Journal, 72, 1370–
1374.
Balesdent J, Chenu C, Balaban M (2000) Relationship of soil organic matter dynamics
to physical protection and tillage. Soil and Tillage Research, 53, 215–230.
Bandyopadhyay PK, Saha S, Mani PK et al. (2010) Effect of organic inputs on aggre-
gate associated organic carbon concentration under long-term rice-wheat cropping
system. Geoderma, 154, 379–386.
Benbi DK, Senapati N (2010) Soil aggregation and carbon and nitrogen stabilization
in relation to residue and manure application in rice-wheat systems in northwest
India. Nutrient Cycling in Agroecosystems, 87, 233–247.
Bhogal A, Nicholson F, Chambers BJ et al. (2009) Organic carbon additions: effects on
soil bio-physical and physic-chemical properties. European Journal of Soil Science,
60, 276–286.
Cai Z, Qin S (2006) Dynamics of crop yields and soil organic carbon in a long-term fer-
tilization experiment in the Huang-Huai-Hai Plain of China. Geoderma, 136, 708–715.
Cai Z, Xing G, Yan X et al. (1997) Methane and nitrous oxide emissions from rice
paddy fields as affected by nitrogen fertilisers and water management. Plant and
Soil, 196, 7–14.
Campbell C, Selles F, Lafond G et al. (1998) Effect of crop management on C and N in
long-term crop rotations after adopting no-tillage management: comparison of soil
sampling strategies. Canadian Journal of Soil Science, 78, 155–162.
Chapin FS III, Matson PA, Mooney HA (2002) Principles of Terrestrial Ecosystem Ecol-
ogy. Springer, New York, NY, USA.
Chen HQ, Hou RX, Gong YS et al. (2009) Effects of 11 years of conservation tillage on
soil organic matter fractions in wheat monoculture in Loess Plateau of China. Soil
and Tillage Research, 106, 85–94.
Cochran VL, Sparrow EB, Schlentner SF et al. (1997) Long-term tillage and crop resi-
due management in the subarctic: fluxes of methane and nitrous oxide. Canadian
Journal of Soil Science, 77, 565–570.
Conen F, Yakutin MV, Sambuu AD (2003) Potential for detecting changes in soil
organic carbon concentrations resulting from climate change. Global Change Biol-
ogy, 9, 1515–1520.
Conrad R, Klose M (2006) Dynamics of the methanogenic archaeal community in
anoxic rice soil upon addition of straw. European Journal of Soil Science, 57, 476–484.
Cui J, Zhang RJ, Bu NS et al. (2012) Changes in soil carbon sequestration and soil res-
piration following afforestation on paddy fields in north subtropical China. Journal
of Plant Ecology, 5, 1–13.
Curtis PS, Wang XZ (1998) A meta-analysis of elevated CO2 effects on woody plant
mass, form, and physiology. Oecologia, 113, 299–313.
Duiker SW, Lal R (1999) Crop residue and tillage effects on carbon sequestration in a
Luvisol in central Ohio. Soil and Tillage Research, 52, 73–81.
Duiker SW, Lal R (2000) Carbon budget study using CO2 flux measurements from a
no till system in central Ohio. Soil and Tillage Research, 54, 21–30.
FAO (2007) A map of Organic carbon (%) in topsoil from digital media series. http://
www.fao.org/fileadmin/templates/nr/images/resources/images/Maps/geonet
work/oc_t.png.
Garcia-Orenes F, Cerda A, Mataix-Solera J et al. (2009) Effects of agricultural manage-
ment on surface soil properties and soil-water losses in eastern Spain. Soil and Till-
age Research, 106, 117–123.
Gurevitch J, Hedges LV (1999) Statistical issues in ecological meta-analyses. Ecology,
80, 1142–1149.
Haynes R (2000) Labile organic matter as an indicator of organic matter quality in ara-
ble and pastoral soils in New Zealand. Soil Biology and Biochemistry, 32, 211–219.
Hedges LV, Gurevitch J, Curtis PS (1999) The meta-analysis of response ratios in
experimental ecology. Ecology, 80, 1150–1156.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
CARBON DYNAMIC AT STRAW RETURN 1379
Huang Y, Liu SL, Shen QR et al. (2002) Influence of environmental factors on the
decomposition of organic carbon in agricultural soils. Chinese Journal of Applied
Ecology, 13, 709–714 (Abstract).
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on Cli-
mate Change (eds Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB,
Tignor M, Miller HL), Cambridge University Press, Cambridge, UK, and New
York, NY, USA.
Iqbal J, Hu RG, Lin S et al. (2009) CO2 emission in a subtropical red paddy soil (Ulti-
sol) as affected by straw and N-fertilizer applications: a case study in southern
China. Agriculture, Ecosystems and Environment, 131, 292–302.
Jabro J, Sainju U, Stevens WB et al. (2008) Carbon dioxide flux as affected by tillage
and irrigation in soil converted from perennial forages to annual crops. Journal of
Environmental Management, 88, 1478–1484.
Jacinthe PA, Lal R, Kimble JM et al. (2002) Effects of wheat residue fertilization on
accumulation and biochemical attributes of organic carbon in a central Ohio luv-
isol. Soil Science, 167, 750–758.
Jarecki MK, Lal R (2003) Crop management for soil carbon sequestration. Critical
Reviews in Plant Sciences, 22, 471–502.
Jastrow JD (1996) Soil aggregate formation and the accrual of particulate and mineral-
associated organic matter. Soil Biology and Biochemistry, 28, 665–676.
Jenkinson D (1971) Studies on the decomposition of 14C labelled organic matter in
soil. Soil Science, 111, 64–70.
Johnson LC, Shaver GR, Cades DH et al. (2000) Plant carbon–nutrient interactions con-
trol CO2 exchange in Alaskan wet sedge tundra ecosystems. Ecology, 81, 453–469.
Khalil K, Mary B, Renault P (2004) Nitrous oxide production by nitrification and deni-
trification in soil aggregates as affected by O2 concentration. Soil Biology and Bio-
chemistry, 36, 687–699.
Kong AYY, Six J, Bryant DC et al. (2005) The relationship between carbon input,
aggregation, and soil organic carbon stabilization in sustainable cropping systems.
Soil Science Society of America Journal, 69, 1078–1085.
Kuzyakov Y, Schneckenberger K (2004) Review of estimation of plant rhizodeposition
and their contribution to soil organic matter formation. Archives of Agronomy and
Soil Science, 50, 115–132.
Lal R (2004a) Offsetting China’s CO2 emissions by soil carbon sequestration. Climatic
Change, 65, 263–275.
Lal R (2004b) Soil carbon sequestration impacts on global climate change and food
security. Science, 304, 1623–1627.
Lal R (2008) Crop Residues and Soil Carbon. Conservation Agriculture Carbon Offset
Consultation (October 28-30, 2008). Food and Agriculture Organization of the Uni-
ted Nations and Conservation Technology Information Center, Beck Agricultural
Center, West Lafayette, Indiana.
Lal R, Bruce JP (1999) The potential of world cropland soils to sequester C and miti-
gate the greenhouse effect. Environmental Science & Policy, 2, 177–185.
Lal R, Kimble JM (1997) Conservation tillage for carbon sequestration. Nutrient
Cycling in Agroecosystem, 49, 243–253.
Le Qu�er�e C, Raupach MR, Canadell JG et al. (2009) Trends in the sources and sinks of
carbon dioxide. Nature Geoscience, 2, 831–836.
Li ZP, LiuM,Wu XC et al. (2010) Effects of long-term chemical fertilization and organic
amendments on dynamics of soil organic C and total N in paddy soil derived from
barren land in subtropical China. Soil and Tillage Research, 106, 268–274.
Liu L, Greaver TL (2009) A review of nitrogen enrichment effects on three biogenic
GHGs: the CO2 sink may be largely offset by stimulated N2O and CH4 emission.
Ecology Letters, 12, 1103–1117.
Lou YS, Li ZP, Zhang TL et al. (2004) CO2 emissions from subtropical arable soils of
China. Soil Biology and Biochemistry, 36, 1835–1842.
Lu F, Wang XK, Han B et al. (2009) Soil carbon sequestrations by nitrogen fertilizer
application, straw return and no-tillage in China’s cropland. Global Change Biology,
15, 281–305.
Lu M, Yang Y, Luo Y et al. (2011a) Responses of ecosystem nitrogen cycle to nitrogen
addition: a meta-analysis. New Phytologist, 189, 1040–1050.
Lu M, Zhou XH, Luo YQ et al. (2011b) Minor stimulation of soil carbon storage by
nitrogen addition: a meta-analysis. Agriculture, Ecosystems and Environment, 140,
234–244.
Lugato E, Berti A, Giardini L (2006) Soil organic carbon (SOC) dynamics with and
without residue incorporation in relation to different nitrogen fertilisation rates.
Geoderma, 135, 315–321.
Luo YQ, Weng ES (2011) Dynamic disequilibrium of the terrestrial carbon cycle under
global change. Trends in Ecology & Evolution, 26, 96–104.
Luo YQ, Zhou XH (2006) Soil Respiration and the Environment. Academic Press/Else-
vier, San Diego, CA.
Luo YQ, Hui DF, Zhang DQ (2006) Elevated CO2 stimulates net accumulations
of carbon and nitrogen in land ecosystems: a meta-analysis. Ecology, 87, 53–
63.
Luo ZK, Wang EL, Sunc OJ (2010) Can no-tillage stimulate carbon sequestration in
agricultural soils? a meta-analysis of paired experiments. Agriculture, Ecosystems
and Environment, 139, 224–231.
Ma J, Xu H, Yagi K et al. (2008) Methane emission from paddy soils as affected by
wheat straw returning mode. Plant and Soil, 313, 167–174.
Malhi SS, Nyborg M, Goddard T et al. (2011) Long-term tillage, straw and N rate
effects on some chemical properties in two contrasting soil types in western
Canada. Nutrient Cycling in Agroecosystems, 90, 133–146.
Naser HM, Nagata O, Tamura S et al. (2007) Methane emissions from five paddy
fields with different amounts of rice straw application in central Hokkaido, Japan.
Soil Science and Plant Nutrition, 53, 95–101.
Niu LA, Hao JM, Zhang BZ et al. (2011) Influences of long-term fertilizer and
tillage management on soil fertility of the north China plain. Pedosphere, 21, 813–
820.
Otte S, Grobben NG, Robertson LA et al. (1996) Nitrous oxide production by Alcalige-
nes faecalis under transient and dynamic aerobic and anaerobic conditions. Applied
and Environmental Microbiology, 62, 2421–2426.
Pan GX, Li LQ, Wu LS et al. (2003) Storage and sequestration potential of topsoil
organic carbon in China’s paddy soils. Global Change Biology, 10, 79–92.
Pan GX, Zhou P, Li ZP et al. (2009) Combined inorganic/organic fertilization
enhances N efficiency and increases rice productivity through organic carbon
accumulation in a rice paddy from the Tai Lake region, China. Agriculture Ecosys-
tems and Environment, 131, 274–280.
Pathak H, Singh R, Bhatia A et al. (2006) Recycling of rice straw to improve wheat
yield and soil fertility and reduce atmospheric pollution. Paddy Water Environment,
4, 111–117.
Paustian K, Six J, Elliott ET et al. (2000) Management options for reducing CO2 emis-
sions from agricultural soils. Biogeochemistry, 48, 147–163.
Rosenberg MS, Adams DC, Gurevitch J (2000) MetaWin: Statistical Software for Meta-
analysis. Sinauer Associates, Sunderland, MA, USA.
Schulz E, Breulmann M, Boettgerb T et al. (2011) Effect of organic matter input on
functional pools of soil organic carbon in a long-term double rice crop experiment
in China. European Journal of Soil Science, 62, 134–143.
Shang Q, Yang X, Gao C et al. (2011) Net annual global warming potential and
greenhouse gas intensity in chinese double rice-cropping systems: a 3-year field
measurement in long-term fertilizer experiments. Global Change Biology, 17,
2196–2210.
Shen MX, Yang LZ, Yao XM et al. (2007) Long-term effects of fertilizer managements
on crop yields and organic carbon storage of a typical rice–wheat agroecosystem
of China. Biology and Fertility of Soils, 44, 187–200.
Shields J, Paul E (1973) Decomposition of 14C-labelled plant material under field con-
ditions. Canadian Journal of Soil Science, 53, 297–306.
Six J, Conant R, Denef K et al. (2002) Stabilization mechanisms of soil organic matter:
implications for C-saturation of soils. Plant and Soil, 241, 155–176.
Six J, Ogle SM, Conant RT et al. (2004) The potential to mitigate global warming with
no-tillage management is only realized when practised in the long term. Global
Change Biology, 10, 155–160.
Smith P, Fang CM (2010) Carbon cycle: a warm response by soils. Nature, 464, 499–
500.
Smith P, Powlson DS, Smith JU et al. (2000) Meeting Europe’s climate change commit-
ments: quantitative estimates of the potential for carbon mitigation by agriculture.
Global Change Biology, 6, 525–539.
Sommer R, Ryan J, Masri S et al. (2011) Effect of shallow tillage, moldboard plow-
ing, straw management and compost addition on soil organic matter and nitro-
gen in a dryland barley/wheat-vetch rotation. Soil and Tillage Research, 115–116,
39–46.
Song G, Li LQ, Pan GX et al. (2005) Topsoil organic carbon storage of China and its
loss by cultivation. Biogeochemistry, 74, 47–62.
Stewart CE, Paustian K, Conant RT et al. (2007) Soil carbon saturation: concept, evi-
dence and evaluation. Biogeochemistry, 86, 19–31.
Triberti L, Nastri A, Giordani G et al. (2008) Can mineral and organic fertilization
help sequestrate carbon dioxide in cropland? European Journal of Agronomy, 29,
13–20.
Vitousek PM, Howarth RW (1991) Nitrogen limitation on land and in the sea: how
can it occur? Biogeochemistry, 13, 87–115.
Wang JB, Chen ZH, Chen LJ et al. (2011) Surface soil phosphorus and phosphatase
activities affected by tillage and crop residue input amounts. Plant Soil and Envi-
ronment, 57, 251–257.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
1380 C. LIU et al.
West TO, Marland G (2002) A synthesis of carbon sequestration, carbon emissions,
and net carbon flux in agriculture: comparing tillage practices in the United States.
Agriculture Ecosystems and Environment, 91, 217–232.
West TO, Six J (2007) Considering the influence of sequestration duration and carbon
saturation on estimates of soil carbon capacity. Climatic Change, 80, 25–41.
Xie ZB, Liu G, Bei QC et al. (2010) CO2 mitigation potential in farmland of China by
altering current organic matter amendment pattern. Science China Earth Sciences,
53, 1351–1357.
Yagi K, Tsuruta H, Kanda K et al. (1996) Effect of water management on methane
emission from a japanese rice paddy field: automated methane monitoring. Global
Biogeochemical Cycles, 10, 255–267.
Zhang W, Yu Y, Huang Y et al. (2011) Modeling methane emissions from irrigated
rice cultivation in China from 1960 to 2050. Global Change Biology, 17, 3511–3523.
Zhou P, Song GH, Pan GX et al. (2009) SOC enhancement in three major types of
paddy soils in a long-term agro-ecosystem experiment in south China. II. chemi-
cal binding and protection in micro-aggregate size fractions. Acta Pedologica Sini-
ca, 46, 263–273.
Zhu HH, Wu JS, Huang DY et al. (2010) Improving fertility and productivity of a
highly-weathered upland soil in subtropical China by incorporating rice straw.
Plant and Soil, 331, 427–437.
Zou JW, Huang Y, Lu YY et al. (2005) Direct emission factor for N2O from rice–winter
wheat rotation systems in southeast China. Atmospheric Environment, 39, 4755–4765.
Supporting Information
Additional Supporting Information may be found in the online version of this article:
Appendix S1. List of studies included in this meta-analysis.Appendix S2. Percentage of change of 23 variables related to agricultural soil C cycles in response to straw-C input.Appendix S3. Statistical results of comparisons among groups.Appendix S4. Relationships between the lnRR of SOC for the response to straw addition of mean annual temperature (a), meanannual precipitation (b), and N application rate (c).Appendix S5. Global distribution of straw return experiments included in this meta-analysis.
© 2014 John Wiley & Sons Ltd, Global Change Biology, 20, 1366–1381
CARBON DYNAMIC AT STRAW RETURN 1381
top related