Content uploaded by Lucas Silva
Author content
All content in this area was uploaded by Lucas Silva
Content may be subject to copyright.
TECHNICAL REPORTS
912
Common management practices, such as the application of
green waste compost, soil moisture manipulation, and nitrogen
fertilization, a ect nitrous oxide (N2O) emissions from agricultural
soils. To expand our understanding of how soils interact with these
controls, we studied their e ects in 10 agricultural soils. Application
of compost slightly increased N2O emissions in soils with low initial
levels of inorganic N and low background emission. For soils in
which compost caused a decrease in emission, this decrease was
larger than any of the observed increases in the other soils. e ve
most important factors driving emission across all soils, in order
of increasing importance, were native dissolved organic carbon
(DOC), treatment-induced change in DOC, native inorganic N,
change in pH, and soil iron (Fe). Notable was the prominence of
Fe as a regulator of N2O emission. In general, compost is a viable
amendment, considering the agronomic bene ts it provides against
the risk of producing a small increase in N2O emissions. However, if
soil properties and conditions are taken into account, management
can recognize the potential e ect of compost and thereby reduce
N2O emissions from susceptible soils, particularly by avoiding
application of compost under wet conditions and together with
ammonium fertilizer.
Quantifying the E ects of Green Waste Compost Application,
Water Content and Nitrogen Fertilization on Nitrous Oxide
Emissions in 10 Agricultural Soils
Xia Zhu,* Lucas C.R. Silva, Timothy A. Doane, Ning Wu, and William R. Horwath
N (N2O) is a key trace gas that plays
a central role in atmospheric geochemistry. Nitrous
oxide is a major greenhouse gas that contributes to
climate warming and depletes stratospheric ozone (Crutzen,
1981). Globally, agricultural soils are the major source of anthro-
pogenic N2O emissions (IPCC, 2007). ere is little doubt that
the increasing atmospheric concentration of N2O is primarily
caused by the use of nitrogen (N) fertilizer (Davidson, 2009).
However, the production of N2O from fertilizer N depends
primarily on soil properties, some of which are in uenced by
management and incorporation of organic amendments, such as
compost (Wright et al., 2008), water manipulation through irri-
gation techniques, and other agronomic practices, such as types
of fertilizer N (Mosier et al., 1998).
e production of N2O in soils occurs through denitri cation
and ammonium oxidation (nitri er nitri cation, nitri er
denitri cation, and nitri cation-coupled denitri cation)
(Wrage et al., 2001). Many soil properties in uence N2O
produced through these pathways in soils, such as soil texture,
organic carbon, pH, N form and availability, and concentration
of certain metals, including iron (Fe) (Bouwman et al., 2002;
Bremner et al., 1980; Stehfest and Bouwman, 2006). Most of
these properties can be a ected by agricultural management, such
as compost application, irrigation, and fertilization (Mulvaney et
al., 1997; Wright et al., 2008). Little is known about the role of
Fe in N2O emission from agricultural soils.
Application of organic amendments such as compost
reduces N2O emissions from soils (Dalal et al., 2010), most
likely due to a reduction in mineral N (Wright et al., 2008),
which is the substrate of denitri cation or nitri cation. In
contrast, N2O emission can increase a er compost application
(Mondini et al., 2007). e addition of compost increases
the content of dissolved organic carbon (DOC) (Wright et
al., 2008), which can stimulate microbial activity and lead to
Abbreviations: DOC, dissolved organic carbon; IN, inorganic nitrogen; ODOC,
original dissolved organic carbon; OFe, original Fe; OIN, original inorganic nitrogen;
OpH, original pH; OTC, original total C; OTN, original total N; ON2O, background
nitrous oxide emission; PLS, partial least squares; WHC, water-holding capacity.
X. Zhu and N. Wu, Chengdu Institute of Biology, Chinese Academy of Sciences,
Chengdu 610041, China; X. Zhu, L.C.R. Silva, T. A. Doane, and W.R. Horwath,
Biogeochemistry and Nutrient Cycling Lab., Dep. of Land, Air and Water Resources,
Univ. of California Davis, CA 95616; X. Zhu, Univ. of Chinese Academy of Sciences,
Beijing 100049, China; N. Wu. International Centre for Integrated Mountain
Development, Kathmandu 32800, Nepal. Assigned to Associate Editor Jan Willem
van Groenigen.
Copyright © American Society of Agronomy, Crop Science Society of America,
and Soil Science Society of America. 5585 Guilford Rd., Madison, WI 53711 USA.
All rights reserved. No part of this periodical may be reproduced or transmitted
in any form or by any means, electronic or mechanical, including photocopying,
recording, or any information storage and retrieval system, without permission in
writing from the publisher.
J. Environ. Qual. 42:912–918 (2013)
doi:10.2134/jeq2012.0445
Supplemental data le is available online for this article.
Received 19 Nov. 2012.
*Corresponding author (zhuxia0207@gmail.com).
Journal of Environmental Quality
WASTE MANAGEMENT
TECHNICAL REPORTS
www.agronomy.org • www.crops.org • www.soils.org 913
increased oxygen consumption, resulting in N2O production
via denitri cation and nitri er denitri cation. Denitri cation
is o en considered to be the main source of N2O emitted
from soils because emissions derived from N application tend
to increase with increasing soil water content (Abbasi and
Adams, 2000; Bateman and Baggs, 2005), although ammonia
oxidation–derived N2O measured in the laboratory has been
observed to increase with increasing soil moisture (Bremner and
Blackmer, 1979). ese divergent processes make it di cult for
modelers to predict N2O emissions. A better understanding of
how these multiple controls a ect edaphic properties and drive
N2O production is needed to guide management practices
with the goal of mitigating emissions, particularly with the use
of amendments applied to address soil productivity.
e objective of this study was to develop a quantitative
understanding of the complexities that occur among multiple
controls of N2O production and the main factors a ecting
its production in highly productive agricultural soils a er
application of organic amendments such as compost. Using 10
di erent soils, we performed laboratory incubations to test the
e ect of compost application, soil water content, N fertilizer
form, and their interactions on N2O emission. We discuss our
results presenting the di erent pathways of N2O production
that predominate in di erent soils, providing an integrated
analysis of the most important factors a ecting emissions
across all studied soils, and conclude with recommendations
for mitigation of N2O emissions based on edaphic traits and
optimal management practices.
Materials and Methods
Soil and Compost Sampling
Soil samples were collected from 10 agricultural elds that
were chosen to provide a diverse set of soils from the main
agricultural production areas in California. All soil samples were
collected before fertilizer application. e location, previous
crop, and soil properties at each site are summarized in Table 1.
Composite soil samples from numerous auger borings from 0
to 15 cm in depth were sieved to 2 mm and refrigerated (4°C)
until the experiment began (sample storage time <7 d ). Compost
samples were collected from yard green waste compost windrows,
which had been composted for more than 8 wk and met the
mandated Process to Further Reduce Pathogens requirement for
soil application (Calrecycle, 2012). e respiration of compost
was 2.8 mg CO2–C g−1 organic matter d−1, and the relative
seedling vigor averaged 100% of control, indicating that the
compost was stable and mature (Calrecycle, 2012). e compost
properties are included in Table 1. Compost samples were nely
(<1 mm) ground in the lab using a small laboratory mill and
refrigerated (4°C) until the experiment began.
Soil Incubation Experiment
Soils were weighed (50 g dry mass) into standard 120-mL
specimen cups and placed in 1-L Mason jars containing a septum
for gas sampling and 2 mL of water to maintain sample soil
moisture throughout the incubation. To simulate management
practices typically performed in the eld, 1.2 g (dry mass) of
ground compost was mixed with soil and incubated at 40% of
water-holding capacity (WHC) for 7 d. Identical noncompost
soil samples were incubated under the same conditions. For each
Table 1. Characteristics of the soils and compost.
Soils Location Classi cation Previous
crop
Soil
texture Sand Clay Silt pH† Inorganic
N DOC‡ Total
C
Total
NFe§
——— % ——— —— mg kg−1 —— — g kg−1 — mg kg−1
S1 Sanger coarse-loamy, mixed, nonacid,
thermic Typic Xerorthents
grape Sandy 61 7 32 4.2 3.6 28 3 0.3 260
M2 Modesto ne-loamy, mixed, superactive,
thermic Typic Argixeroll
grape Sandy 72 10 18 6.9 129.9 164 9.7 1.1 240
S3 Salinas ne, montmorillonitic, thermic
Pachic Argixeroll
lettuce Sandy 64 13 23 7.2 5.3 44.0 6.6 0.7 150
C4 Castroville ne, montmorillonitic, thermic
Ultic Palexerol
tomato Sandy 72 13 15 6.4 31.8 88.3 7.5 0.8 550
S5 Spence ne-loamy, mixed, thermic
Typic Argixeroll
artichoke loam 50 21 29 6.6 18.1 63.2 12.8 1.1 270
D6 Davis ne, montmorillonitic, thermic
Mollic Haploxeralfs
tomato loam 30 24 42 5.4 2.2 16.8 8.5 0.9 170
D7 Dixon ne-silty, mixed, nonacid,
thermic Typic Xerorthent
wheat Clay
loam
23 28 49 5.6 10.7 30.4 16 1.4 290
F8 Five Points ne-loamy, mixed, superactive,
thermic Typic Haplocambid
tomato Clay
loam
36 32 32 6.8 3.7 57.2 6.7 0.8 60
S9 Salinas Clear Lake clay: ne,
montmorillonitic thermic typic
pelloxerert
broccoli Clay 22 42 36 7.4 27.7 87.7 17.8 1.6 240
D10 Dixon ne-silty, mixed, nonacid,
thermic Typic Xerorthents
radish Silty clay 15 44 41 5.5 4.6 19.3 11.8 1.1 210
CV % 49 56 34 16 141 77 46 38 52
Compost 8.1 324 270 227 15 1700
† pH measured in KCl 1:1.
‡ Dissolved organic carbon.
§ Pyrophosphate extractable iron.
914 Journal of Environmental Quality
soil compost treatment (compost and noncompost), three N
fertilization treatments were applied: (NH4)2SO4, KNO3, and no
fertilizer (control). Fertilizer treatments received a dose of 100 mg
N kg−1 soil. To assure uniform distribution, fertilization treatments
were applied in water solution and sprayed onto the di erent soils
to obtain a nal moisture content of 50 or 100% of soil WHC. e
no-fertilizer N treatment received deionized water to reproduce
the moisture contents of the treatment samples. e experimental
design and treatment application were set up as completely
randomized blocks with three replicates per treatment (360 total
experimental units) and incubated for 14 d at 22°C. Soil moisture
remained constant throughout the incubation period.
Nitrous Oxide Sampling and Analysis
Gas samples for N2O were taken from the Mason jars on Days
0, 1, 2, 3, 5, 9, and 14 a er preincubation. Jars were closed with
septum-equipped lids for 60 min, and 20-mL gas samples were
removed from each jar headspace at 0, 30, and 60 min a er jar
closure and transferred to 12-mL evacuated vials. e ux was
determined by linear interpolation between samples. Cumulative
N2O–N was calculated using total (daily) estimates of N2O ux,
with an assumption that N2O ux measured on a sampling date
was representative of the average daily ux. Gas samples were
analyzed using a gas chromatograph with an electron capture
detector (Model 2014, Shimadzu Scienti c Instruments).
Soil Extraction and Analysis
At the end of the incubation period, soils were extracted
with 0.5 mol L−1 K
2SO4 (1:10 ratio, w:v). e extracts were
ltered, and the supernatant was stored at 4°C. Within 48h
a er extraction of inorganic N (IN) (NH4
+ and NO3
−),
samples were analyzed using colorimetric methods (Doane
and Horwath, 2003; Verdouw et al., 1978). Dissolved organic
C levels were determined by UV-persulfate oxidation (Phoenix
8000, Teledyne-Tekmar). Soil pH was determined in 1 mol L−1
KCl extracts (1:1, w:v). Total C and N content were determined
using an elemental analyzer (EAS 4010, Costech) a er sample
ball milling preparation. Percent clay, silt, and sand were
determined by a modi ed pipet method (USDA, 1992). Iron
was determined by shaking 1 g soil with 100 mL 0.1 mol L−1
sodium pyrophosphate for 16 h, followed by centrifugation for
30 min at 15,600 × g; further centrifugation did not result in any
di erence in measured Fe concentration, indicating that all ne
Fe colloids had been removed, an important consideration when
using this extraction (Loveland and Digby, 1984).
Statistical Analyses
We analyzed 10 di erent soils with highly variable properties.
To illustrate intrinsic di erences based on the original condition
of these soils, we present the mean values of all measured
properties and the coe cient of variation calculated by dividing
standard deviations by means across all soils (Table 1). e
e ect of treatments (compost application, water content, and N
fertilization) and their interactions with soils on the cumulative
emission of N2O were analyzed by full factorial ANOVA
(Table 2) using log-transformed data to improve variance
homogeneity. Post hoc Tukey’s honestly signi cant di erence
multiple comparisons of means or paired t tests were used when
appropriate to verify signi cant di erences (P < 0.05) between
treatments within a given soil. To convert the e ect of treatments
(categorical variables) into continuous numeric responses, we
calculated change (Δ), comparing measurements from before
and a er treatment and using variables that we expected to
change (pH, DOC, and IN). ese measures of change were
used in partial least squares (PLS) multivariate analysis, which
incorporated nine other continuous numeric variables (soil
original pH [OpH], soil original DOC [ODOC], soil original
IN [OIN], soil original Fe [OFe], clay, sand, silt, soil original
total N [OTN], and soil original total C [OTC]) representing
original edaphic traits that could a ect the N2O emission
potential of soils. e objective of this analysis was to identify
the most important drivers of N2O emissions while accounting
for variability across soils. Background N2O emissions, measured
in control treatments in each soil type, were included in this
analysis as an integrated measure of intrinsic emission potential.
Partial least squares analysis bears some relation to principal
component analysis, representing a form of structural equation
modeling, distinguished from the principal component analysis
method by being component based rather than covariance based
(Esposito Vinzi, 2010; Tenenhaus et al., 2005). e essential
Table 2. E ects of soil, compost, water content, and nitrogen fertilization on cumulative nitrous oxide emissions.
Source DF Sum of squares F ratio Prob > F
Soil 9 70.85 9.46 <0.0001*
Compost 1 9.35 11.24 0.0009*
Water content 1 14.18 17.05 <0.0001*
N fertilization 2 98.34 59.11 <0.0001*
Soil × Compost 9 6.08 0.81 0.605
Soil × Water content 9 27.33 3.65 0.0003*
Soil × N fertilization 18 23.8 1.59 0.063
Soil × Compost × Water content 9 11.64 1.56 0.130
Soil × Compost × N fertilization 18 25.45 1.70 0.040*
Soil × Water × N fertilization 18 12.61 0.84 0.649
Compost × Water content 1 1.15 1.39 0.240
Compost × N fertilization 2 2.12 1.28 0.281
Water content × N fertilization 2 1.67 1.01 0.368
Compost × Water content × N fertilization 2 1.28 0.77 0.466
Soil × Compost × Water content × N fertilization 18 14.35 0.96 0.509
* Signi cant at the 0.05 probability level.
www.agronomy.org • www.crops.org • www.soils.org 915
output of the analysis ranks the most important components
(independent variables) based on linear regression models that
project the predicting variables and the observable variables
to a new (multivariate) space. Partial least squares analysis
is particularly suited when the matrix of predictors (here
represented by original soil properties or Δ a er treatments) has
more variables than the resulting observations (here represented
by N2O emissions) and when multicollinearity is expected among
predictors (Tenenhaus et al., 2005), which is the case here. Partial
least squares is also generally preferable in circumstances where
assumptions of multivariate normality cannot be made but data
can be standardized (Esposito Vinzi, 2010). In our analysis both
predictors and response were standardized (centered and scaled
to have mean 0 and SD 1) before PLS analysis to ensure that the
criterion for choosing the most important factors driving N2O
emission is based on how much variation they explain when
having the same weight. Data standardization and statistical
analyses were performed using JMP 10 so ware (Sall et al., 2005).
Results
The E ect of Treatments on Nitrous Oxide Emissions
Table 2 shows that compost application, water content, and
N fertilization signi cantly a ected N2O emissions, with no
interactions observed among the applied treatments. However,
signi cant interactions occurred between treatments and soils.
In Fig. 1, we show how treatments a ected cumulative N2O
emissions in each soil studied. In soils M2, C4, and D7, compost
application decreased N2O emission, and the decrease in N2O
emission was equivalent to 0.52, 1.4, and 0.61% of the total IN
present at the start of the incubation, respectively (Fig. 1A).
In soils S1, S3, F8, and D10, compost application signi cantly
increased N2O emissions, but this increase was slight compared
with the above-mentioned decrease. In these soils, the increase in
N2O emission due to the compost application was equivalent to
0.05, 0.26, 0.41, and 0.12% conversion of the total IN present at
the start of the incubation, respectively. With the exception of
soils S1, S5, and S9, much higher N2O emissions were observed
with increased water availability (100 vs. 50% WHC) (Fig. 1B).
In all soils except in S5 and D7, the application of (NH4)2SO4
signi cantly increased N2O emissions, whereas no signi cant
e ects of KNO3 application were observed compared with
nonfertilized soil (Fig. 1C). e highest N2O emission occurred
in soil fertilized with (NH4)2SO4 under high moisture content
(100% WHC) (Fig. 1B, C). e dynamics of N2O emission are
shown in the Supplemental Fig. S1.
Original Soil Properties and Change in Response
to Applied Treatments
e original edaphic (pretreatment) properties vary widely
with soil (Table 1). e most variable soil properties include
clay, IN, and DOC. Among the soil variables, pH, DOC, and
IN could change by an unknown amount a er application of
treatments; therefore, these variables were measured again a er
the incubation. Changes in these variables during the incubation
could, in turn, in uence N2O emissions. A full factorial analysis
testing the e ect of compost application, water content, N
fertilization, and the interactions among them on soil pH, DOC,
and IN shows that all treatments signi cantly a ected these
variables, but no signi cant interactions (with the exception of
the water × compost e ect on DOC) were observed among
treatments (Table 3). e application of compost signi cantly
increased soil pH and DOC across all soils, but soil pH decreased
a er (NH4)2SO4 application, and lower pH was observed in
50% WHC compared with 100% WHC. Applied N fertilizer,
regardless of its form, signi cantly increased levels of IN in soils.
A larger increase in IN was found under drier conditions (50%
WHC), but compost application signi cantly decreased total
IN concentration. Soil IN at the end of the incubation is given in
the Supplemental Table S1.
Key Soil Properties Driving Nitrous Oxide Emissions
e PLS multivariate analysis generated a variable importance
plot that describes the relative importance of each measured
variable in explaining total N2O emission across all studied soils.
e contribution of each variable is compared with a signi cance
coe cient (0.8) (Wold, 1995) represented by the black line in
Fig. 2. Any variable in which the explanatory power represented
by the variable importance plot value goes above the black line
is considered signi cant in the multivariate PLS model. In our
analysis, background N2O emission (ON2O), levels of OTN, and
ΔIN were not considered signi cant in explaining emissions. All
other variables (OpH, ODOC, OIN, OTC, clay, silt, sand, OFe,
ΔpH, and ΔDOC) were signi cant. For simplicity, we focus
our discussion on ve variables: OFe, ΔpH, OIN, ΔDOC, and
Fig. 1. Mean N2O emission in compost (A), water content (B), and
N fertilization (C) treatments within soils. *Signi cant di erence
between treatments. Di erent letters indicate a signi cant di erence
among treatments. WHC, water-holding capacity.
916 Journal of Environmental Quality
ODOC. ese variables collectively explain over 75% of the
variation in N2O emissions across soils.
Discussion
The E ect of Treatments on Nitrous Oxide Emissions
Our results show that compost application promoted N2O
emissions in soils S1, S3 (sandy loam), F8 (clay loam), and
D10 (silty clay), suggesting that, in soil with low OIN (<6 μg
g−1) and low background N2O emission (<79 ng g−1), compost
application promotes N2O emission. Nitrous oxide emissions
from soil are mainly the result of ammonium oxidation (nitri er
nitri cation, nitri er denitri cation, and nitri cation coupled
denitri cation) and denitri cation (Wrage et al., 2001).
Increased N2O emissions in soils with low OIN likely occurred as
a result of the compost eliminating N and C substrate limitation
for ammonium oxidation and denitri cation. Decreased N2O
emissions by compost application compared with no compost
were observed in soils M2, C4 (sandy loam), and D7 (clay
loam). e decrease in N2O was signi cantly larger than the
increase in N2O due to compost addition in S1, S3, F8, and D10
soils (Fig. 1A), indicating that it is not possible to draw general
conclusions based solely on soil texture. Compost application is
known to have a substantial e ect on N2O production pathways
(Alluvione et al., 2010; Dalal et al., 2010). e application
of compost in uences several edaphic factors that control
N2O emission pathways, such as soil organic C availability,
N availability, microbial activity, soil pH, and metal content
(Inubushi et al., 2000; Wright et al., 2008). In our study, the
application of compost increased soil organic C availability (i.e.,
increased DOC), which could stimulate soil respiration and
consume O2 in soil (Contreras-Ramos et al., 2009), promoting
N2O emission from nitri er denitri cation and denitri cation
(Bollmann and Conrad, 1998; Wrage et al., 2001). However,
it is well known that a decrease in soil O2 also favors reduction
of N2O to N2 (Firestone et al., 1980) and that increasing soil C
availability stimulates IN assimilation by soil microorganisms.
As a result, less N2O might be emitted from some soils
amended with compost, despite increased nitri er
denitri cation. Our results show that depending on
the interplay of these controlling factors, divergent
e ects can be expected from compost application on
N2O emission.
Contrasting with compost application, the
signi cant increase in N2O emissions with increased
soil moisture is consistent across soils, suggesting
that the process of N2O emission is controlled by O2
di usion in soil and by the availability of substrates
for microbial activity (Stark and Firestone, 1995).
Traditionally, it has been assumed that with increasing
soil moisture and restricted O2 di usion into the
soil, anoxic microbial activity is increased (Renault
and Sierra, 1994), and, as a result, emissions of N2O
derive mostly (or exclusively) from microbial-driven
denitri cation (Abbasi and Adams, 2000). However,
we observed consistent and signi cantly greater N2O
emissions in soils fertilized with (NH4)2SO4 than in
soils fertilized with KNO3, even under high moisture
content (100% WHC) (increase ranging from 3- to
44-fold) (Fig. 1B, C). ese results indicate that ammonium
oxidation, rather than denitri cation, is the predominant process
of N2O production in soil with limited O2 availability. is
Table 3. The e ect of compost, water content, and nitrogen fertilization
and their interaction on soil pH, dissolved organic carbon, and
inorganic nitrogen.
Treatment ΔpH† ΔDOC‡ ΔIN§
———— μg g−1 ————
Compost
No compost −0.08 ± 0.00 0.41 ± 0.24 58.09 ± 0.26
Compost 0.47 ± 0.00 74.24 ± 0.14 49.68 ± 0.26
Water content
50% WHC¶ 0.10 ± 0.00 29.49 ± 0.24 62.07 ± 0.26
100% WHC 0.22 ± 0.00 41.84 ± 0.14 50.33 ± 0.26
N fertilization
Control 0.32 ± 0.00 38.17 ± 0.09 −2.31 ± 0.08
(NH4)2SO4−0.16 ± 0.00 32.35 ± 0.14 79.81 ± 0.14
KNO30.33 ± 0.00 36.47 ± 0.10 91.09 ± 0.13
Compost *** *** ***
Water content ** *** ***
N fertilization *** NS# ***
Compost × Water content NS * NS
Compost × N fertilization NS NS NS
Water content × N
fertilization NS NS NS
Compost × Water content
× N fertilization NS NS NS
* Signi cant at the 0.05 probability level.
** Signi cant at the 0.01 probability level.
*** Signi cant at the 0.001 probability level.
† Change in pH. Change (Δ) measured as the di erence between
posttreatment values and original conditions across all studied soils
(see Table 1).
‡ Change in dissolved organic carbon.
§ Change in inorganic nitrogen.
¶ Water-holding capacity.
# Not signi cant.
Fig. 2. Results from partial least squares analysis. The relative importance of each
measured variable in explaining total N2O emission across all studied soil types.
ΔDOC, change in dissolved organic carbon; ΔIN, change in inorganic nitrogen;
ΔpH, change in pH; ODOC, soil original DOC; OFe, soil original Fe; OIN, soil original
inorganic nitrogen; ON2O, soil original nitrous oxide; OTC, soil original total carbon;
OpH, soil original pH; OTN, soil original total nitrogen.
www.agronomy.org • www.crops.org • www.soils.org 917
notion has been previously supported by Bremner and Blackmer
(1979) and can be explained by nitri er denitri cation, a process
promoted by ammonia oxidizing bacteria, which uses NO2
−
instead of O2 as a terminal electron acceptor, reducing NO2
− to
N2O (Goreau et al., 1980; Poth and Focht, 1985).
Original Soil Properties and Change in Response
to Applied Treatments
e soils used in this study were collected from di erent
agricultural areas under di erent climatic and geomorphological
conditions. erefore, intrinsic properties related to soil genesis
and development, such as texture, Fe content, and levels of DOC
and IN, vary widely with soil type. Because the compost pH
(8.1) was higher than soil pH (range, 4.2–7.4) in our study, soil
pH was signi cantly increased by compost application (Table
3) across all soils. Lower soil pH was observed in 50% WHC
compared with in 100% WHC treatment, which suggests that
nitri cation is stronger in drier conditions, as nitri cation
contributes to increases soil acidity (reduces pH) (Mulvaney et
al., 1997). Soil IN and DOC were also signi cantly in uenced
by compost application, water content, and N fertilization. e
application of compost not only added DOC but also added IN
to soils (Table 1). However, IN content at the end of incubation
was lower in the compost treatments compared with the
no-compost treatments. is can be explained by two factors:
(i) the application of compost promoted N gas loss due to rapid
consumption of oxygen by microbial respiration (Vaughan et
al., 2011) and caused an increase in soil pH (Venterea, 2007),
and (ii) more added IN was immobilized by microbial biomass
as more DOC was available (Hood et al., 2000; Seligman et
al., 1986). Less IN was found a er incubation at 100% WHC
compared with 50% WHC, probably because greater N gas loss
occurred in wet soil than in dry soil (Fig. 1B).
Key Soil Properties Driving Nitrous Oxide Emissions
Our results show that the in uence of treatments on total
N2O emissions varied in di erent soils. Inherent soil properties
combined with the properties that changed a er treatments
contributed to these results. Following the dynamics of certain
key variables can aid in explaining mechanisms of N2O generation
under a variety of conditions. Rather than use a traditional time
course of measurements, which would involve many data points,
our approach to “follow the dynamics” of certain variables was
to quantify how these variables changed during the experiment
as an indicator of a soil’s response to speci c conditions. e
measured values of these responses were included as variables
alongside initial soil properties in the analysis of relative
predictive ability. Some of these treatment-induced changes in
properties were important in explaining N2O emissions (Fig.
2). From all measured parameters integrated across all soils,
the main variables a ecting N2O emission were, in increasing
order of importance, ODOC < ΔDOC < OIN < ΔpH <
OFe. Typically, the variables described as major regulators in
N2O emission from soils are soil texture, soil organic C, soil pH,
and N form and availability (Bouwman et al., 2002; Stehfest
and Bouwman, 2006). Although our results support that these
variables signi cantly a ect N2O emissions, OFe (traditionally
omitted from analysis of N2O emission in agricultural soils)
appears as the most important soil variable a ecting N2O
emissions. e result from linear regression analysis shows
that the amount of OFe was signi cantly positively correlated
with N2O emission (slope = 0.53; F ratio = 4.60; P = 0.03).
e impact of Fe (as pyrophosphate-extractable Fe, which
represents iron complexed with soil organic matter) can be
explained by abiotic and biotic potential mechanisms. On the
abiotic aspect, (i) the reaction of ferric iron (Fe3+) with reactive
nitri cation intermediates such as hydroxylamine can yield
N2O or N2 as the oxidation product depending on the relative
concentration of Fe3+ to hydroxylamine (Bengtsson et al., 2002),
and (ii)chemical nitrite reduction by ferrous iron (Fe2+) can also
yield N2O (Kampschreur et al., 2011). On the biotic aspect, it is
widely known that Fe is contained in metalloenzymes involved
in N2O production (Glass and Orphan, 2012). e ability
of microorganisms to acquire iron consequently in uences
the amount of N2O produced. e application of compost
increased soil pH, which could have the potential to decrease the
availability of Fe. However, compost also brought Fe into soil (34
μg g−1 dry soil), even though the same amount was added across
all soils. In either case, the OFe content of the soils studied here
appears to be a more important regulator of N2O emission than
other properties traditionally considered as the main regulators
of N2O production in soil (e.g., pH, DOC, texture); thus, the
role of Fe should be further investigated.
Regarding the other most important variables, DOC, which
here was increased by compost application and water content in
all soils, could increase N2O emissions by stimulating microbial
activity (Fortuna et al., 2012), by limiting oxygen availability
for nitri er and denitri ers (Coyne and Tiedje, 1990; Goreau
et al., 1980), by decreasing N2O emissions by promoting IN
immobilization by microbial biomass (Hood et al., 2000;
Seligman et al., 1986), or by increasing the reduction rate of
N2O to N2 (Firestone et al., 1980). Indigenous IN was a more
important factor driving N2O emission than the increase in
total IN caused by fertilization treatments. At the end of the
incubation, there was no signi cant di erence in soil IN between
the (NH4)2SO4 and KNO3 treatments, whereas N2O emission
was signi cantly increased by the application of (NH4)2SO4
compared with KNO3.
Conclusions
Here we show that compost application, water content, and
N fertilization have important e ects on N2O emissions from
agricultural soils. e application of compost increased N2O
emission in soils with low content of OIN and low background
N2O emission. For soils in which the compost application caused
a decrease in N2O emission, this decrease was larger than any of
the increases caused by compost application. In soils fertilized
with (NH4)2SO4, N2O emissions were greater than in soils
fertilized with KNO3, even under 100% WHC (from 3- to
44-fold increase). is indicates that ammonium oxidation is the
main pathway of N2O production under limited O2 availability.
Soil pH and DOC were increased, whereas IN was decreased by
compost application at the end of incubation. Among all studied
variables, extractable Fe was the most important factor regulating
soil N2O emission, followed by ΔpH, OIN, ΔDOC, and ODOC.
ese results indicate that to mitigate N2O emissions, compost
application should be avoided in high soil moisture conditions
918 Journal of Environmental Quality
and in combination with NH4
+ fertilization events in soils with
low background N2O emission. e role of Fe as a regulator of
N2O production in soils should be further investigated.
Acknowledgments
We gratefully acknowledge the assistance of those who collected the
soil samples. Support was provided by the California Department
of Resources Recycling and Recovery Agreement IWM09027, the
J.G. Boswell Endowed Chair in Soil Science, and the University of
California Agricultural Experiment Station.
References
Abbasi, M.K., and W.A. Adams. 2000. Gaseous N emission during simultaneous
nitri cation-denitri cation associated with mineral N fertilization to a
grassland soil under eld conditions. Soil Biol. Biochem. 32:1251–1259.
doi:10.1016/S0038-0717(00)00042-0
Alluvione, F., C. Bertora, L. Zavattaro, and C. Grignani. 2010. Nitrous oxide and
carbon dioxide emissions following green manure and compost fertilization
in corn. Soil Sci. Soc. Am. J. 74:384–395. doi:10.2136/sssaj2009.0092
Bateman, E.J., and E.M. Baggs. 2005. Contributions of nitri cation and
denitri cation to N2O emissions from soils at di erent water- lled pore
space. Biol. Fertil. Soils 41:379–388. doi:10.1007/s00374-005-0858-3
Bengtsson, G., S. Fronaeus, and L. Bengtsson-Kloo. 2002. e kinetics and
mechanism of oxidation of hydroxylamine by iron(III). J. Chem. Soc.
Dalton Trans. 12:2548–2552. doi:10.1039/b201602h
Bollmann, A., and R. Conrad. 1998. In uence of O2 availability on NO and
N2O release by nitri cation and denitri cation in soils. Glob. Change
Biol. 4:387–396. doi:10.1046/j.1365-2486.1998.00161.x
Bouwman, A.F., L.J.M. Boumans, and N.H. Batjes. 2002. Modeling global
annual N2O and NO emissions from fertilized elds. Global Biogeochem.
Cycles 16(4):1080. doi:10.1029/2001GB001812
Bremner, J.M., and A.M. Blackmer. 1979. E ects of acetylene and soil-water
content on emission of nitrous oxide from soils. Nature 280:380–381.
doi:10.1038/280380a0
Bremner, J.M., A.M. Blackmer, and S.A. Waring. 1980. Formation of nitrous
oxide and dinitrogen by chemical decomposition of hydroxylamine in soils.
Soil Biol. Biochem. 12:263–269. doi:10.1016/0038-0717(80)90072-3
Calrecycle. 2012. Compostable materials handling operations and facilities
regulatory requirements. http://www.calrecycle.ca.gov/Laws/Reg ulations/
title14/ch31.htm (accessed 20 May 2012).
Contreras-Ramos, S.M., D. Alvarez-Bernal, J.A. Montes-Molina, O. Van
Cleemput, and L. Dendooven. 2009. Emission of nitrous oxide from
hydrocarbon contaminated soil amended with waste water sludge and
earthworms. Appl. Soil Ecol. 41:69–76. doi:10.1016/j.apsoil.2008.09.001
Coyne, M.S., and J.M. Tiedje. 1990. Induction of denitrifying enzymes in
oxygen-limited achromobacter-cycloclastes continuous culture. FEMS
Microbiol. Ecol. 73:263–270. doi:10.1111/j.1574-6968.1990.tb03949.x
Crutzen, P.J. 1981. Atmospheric chemical processes of the oxides of nitrogen including
nitrous oxide. In: C.C. Delwiche, editors, Denitri cation, nitri cation, and
atmospheric nitrous oxide. John Wiley & Sons, New York. p. 17–44.
Dalal, R.C., I. Gibson, D.E. Allen, and N.W. Menzies. 2010. Green waste compost
reduces nitrous oxide emissions from feedlot manure applied to soil. Agric.
Ecosyst. Environ. 136:273–281. doi:10.1016/j.agee.2009.06.010
Davidson, E.A. 2009. e contribution of manure and fertilizer nitrogen
to atmospheric nitrous oxide since 1860. Nat. Geosci. 2:659–662.
doi:10.1038/ngeo608
Doane, T.A., and W.R. Horwath. 2003. Spectrophotometric determination of
nitrate with a single reagent. Anal. Lett. 36:2713–2722. doi:10.1081/
AL-120024647
Esposito Vinzi, V. 2010. Handbook of partial least squares: Concepts, methods
and applications. Springer, New York.
Firestone, M.K., R.B. Firestone, and J.M. Tiedje. 1980. Nitrous oxide from
soil denitri cation- factors controlling its biological production. Science
208:749–751. doi:10.1126/science.208.4445.749
Fortuna, A.M., C.W. Honeycutt, G. Vandemark, T.S. Gri n, R.P. Larkin,
Z.Q. He, B.J. Wienhold, K.R. Sistani, S.L. Albrecht, B.L. Woodbury,
H.A. Torbert, J.M. Powell, R.K. Hubbard, R.A. Eigenberg, R .J. Wright,
J.R. Alldredge, and J.B. Harsh. 2012. Links among nitri cation, nitri er
communities, and edaphic properties in contrasting soils receiving dairy
slurry. J. Environ. ual. 41:262–272. doi:10.2134/jeq2011.0202
Glass, J.B., and V.J. Orphan. 2012. Trace metal requirements for microbial
enzymes involved in the production and consumption of methane and
nitrous oxide. Front. Microbiol. 3:61.
Goreau, T.J., W.A. Kaplan, S.C. Wofsy, M.B. Mcelroy, F.W. Valois, and S.W.
Watson. 1980. Production of NO2- and N2O by nitrifying bacteria at
reduced concentrations of oxygen. Appl. Environ. Microbiol. 40:526–532.
Hood, R., R. Merckx, E.S. Jensen, D. Powlson, M. Matijevic, and G. Hardarson.
2000. Estimating crop N uptake from organic residues using a new
approach to the 15N isotope dilution technique. Plant Soil 223:33–44.
doi:10.1023/A:1004789103949
Inubushi, K., S. Goyal, K. Sakamoto, Y. Wada, K. Yamakawa, and T. Arai. 2000.
In uences of application of sewage sludge compost on N2O production in
soils. Chemosphere: Global Change Sci. 2:329–334.
IPCC. 2007. Mitigation of climate change. In: B. Metz et al., editors,
Contribution of working group III to the fourth assessment report of
the intergovernmental panel on climate change, Cambridge Univ Press,
Cambridge, UK. p. 105.
Kampschreur, M.J., R. Kleerebezem, W.W.J.M. de Vet, and M.C.M. van
Loosdrecht. 2011. Reduced iron induced nitric oxide and nitrous oxide
emission. Water Res. 45:5945–5952. doi:10.1016/j.watres.2011.08.056
Loveland, P.J., and P. Digby. 1984. e extraction of Fe and Al by 0.1
M-pyrophosphate solutions: A comparison of some techniques. J. Soil Sci.
35:243–250. doi:10.1111/j.1365-2389.1984.tb00280.x
Mondini, C., M.L. Cayuela, T. Sinicco, F. Cordaro, A. Roig, and M.A. Sanchez-
Monedero. 2007. Greenhouse gas emissions and carbon sink capacity of
amended soils evaluated under laboratory conditions. Soil Biol. Biochem.
39:1366–1374. doi:10.1016/j.soilbio.2006.12.013
Mosier, A.R., J.A. Delgado, and M. Keller. 1998. Methane and nitrous oxide
uxes in an acid oxisol in western Puerto Rico: E ects of tillage, liming
and fertilization. Soil Biol. Biochem. 30:2087–2098. doi:10.1016/
S0038-0717(98)00085-6
Mulvaney, R.L., S.A. Khan, and C.S. Mulvaney. 1997. Nitrogen fertilizers
promote denitri cation. Biol. Fertil. Soils 24:211–220. doi:10.1007/
s003740050233
Poth, M., and D.D. Focht. 1985. 15N Kinetic analysis of N2O production by
nitrosomonas-europaea: An examination of nitri er denitri cation. Appl.
Environ. Microbiol. 49:1134–1141.
Renault, P., and J. Sierra. 1994. Modeling oxygen di usion in aggregated soils:
2. Anaerobiosis in topsoil layers. Soil Sci. Soc. Am. J. 58:1023–1030.
doi:10.2136/sssaj1994.03615995005800040005x
Sall, J., L. Creighton, and A. Lehman. 2005. JMP start statistics: A guide to
statistics and data analysis using JMP and JMP in so ware. omson
Brooks/Cole, Paci c Grove, CA.
Seligman, N.G., S. Feigenbaum, D. Feinerman, and R.W. Benjamin. 1986.
Uptake of nitrogen from high C-to-N ratio, 15N labeled organic residues
by spring wheat grown under semiarid conditions. Soil Biol. Biochem.
18:303–307. doi:10.1016/0038-0717(86)90065-9
Stark, J.M., and M.K. Firestone. 1995. Mechanisms for soil moisture e ects on
activity of nitrifying bacteria. Appl. Environ. Microbiol. 61:218–221.
Stehfest, E., and L. Bouwman. 2006. N2O and NO emission from agricultural
elds and soils under natural vegetation: Summarizing available
measurement data and modeling of global annual emissions. Nutr. Cycling
Agroecosyst. 74:207–228. doi:10.1007/s10705-006-9000-7
Tenenhaus, M., V.E. Vinzi, Y.M. Chatelin, and C. Lauro. 2005. PLS path
modeling. Comput. Stat. Data Anal. 48:159–205. doi:10.1016/j.
csda.2004.03.005
USDA. 1992. Soil survey laboratory method manual. Soil survey Investigations
Rep. 42:8. USDA, Washington, DC.
Vaughan, S.M., R.C. Dalal, S.M. Harper, and N.W. Menzies. 2011. E ect of
fresh green waste and green waste compost on mineral nitrogen, nitrous
oxide and carbon dioxide from a vertisol. Waste Manag. 31:1720–1728.
doi:10.1016/j.wasman.2011.03.019
Venterea, R.T. 2007. Nitrite-driven nitrous oxide production under aerobic
soil conditions: Kinetics and biochemical controls. Glob. Change Biol.
13:1798–1809. doi:10.1111/j.1365-2486.2007.01389.x
Verdouw, H., C.J.A. Vanechteld, and E.M.J. Dekkers. 1978. Ammonia
determination based on indophenol formation with sodium salicylate.
Water Res. 12:399–402. doi:10.1016/0043-1354(78)90107-0
Wold, S. 1995. Chemometrics: What do we mean with it, and what
do we want from it? Chemom. Intell. Lab. Syst. 30:109–115.
doi:10.1016/0169-7439(95)00042-9
Wrage, N., G.L. Velthof, M.L. van Beusichem, and O. Oenema. 2001. Role
of nitri er denitri cation in the production of nitrous oxide. Soil Biol.
Biochem. 33:1723–1732. doi:10.1016/S0038-0717(01)00096-7
Wright, A.L., T.L. Provin, F.M. Hon, D.A. Zuberer, and R.H. White. 2008.
Compost impacts on dissolved organic carbon and available nitrogen and
phosphorus in turfgrass soil. Waste Manag. 28:1057–1063. doi:10.1016/j.
wasman.2007.04.003
