Content uploaded by Emilia Urbanek
Author content
All content in this area was uploaded by Emilia Urbanek on Sep 16, 2014
Content may be subject to copyright.
METHANOGENIC POTENTIAL OF ARCHIVED SOILS
Małgorzata BRZEZIŃSKA1, Emilia URBANEK2, Paweł SZARLIP1,
Teresa WŁODARCZYK1, Piotr BULAK1, Anna WALKIEWICZ1
& Paweł RAFALSKI1
1 Institute of Agrophysics, Polish Academy of Sciences, 20-290 Lublin, Doświadczalna 4, Poland;
m.brzezinska@ipan.lublin.pl
2 Department of Geography, Swansea University, Singleton Park, SA2 8PP, Swansea, UK,
E.Urbanek@swansea.ac.uk
Abstract: Methane (CH4) is an important element of the biogeochemical carbon cycle. Methanogenic
Archaea are strict anaerobes able to survive in dry and oxic soils, but not in liquid or agar slurry. Little is
known about the mechanisms of their survival. The aim of this paper is to study the methanogenic
potential of mineral soils stored as air-dry over 20 years. We tested the hypothesis that the recovery of
CH4 formation is strongly associated with soil textures. Samples of 16 mineral topsoils characterized by
various Corg, pH and particle size distribution (PSD), and stored under air-dry conditions over 20 years
were flooded with: i) water and incubated in N2 atmosphere, or ii) with glucose solution without
headspace gas exchange and incubated for 132 days. Gases were measured chromatographically, PSD by
laser diffraction method. Microbial activity was restored in all tested soils, and CH4 and CO2 production
started within a few days or weeks after flooding, depending on soil properties and incubation conditions.
The glucose amendment resulted in a 2.8-fold increase in the total CH4 and CO2 release. However, in the
presence of glucose, methanogens in three soils were outcompeted by other microorganisms, and required a
long 132-d lag phase or did not start CH4 production at all. The CH4 positively correlated with the finer soil
fractions, especially with fine silt and clay, while negatively with medium and coarse sand fractions.
Consequently, silt loam soils showed approximately 5 and 2.5 times higher CH4 production, than soils with
coarser textures (sand and sandy loam soils, respectively). In contrast, CO2 production was not influenced by
soil texture. The Corg and moisture retention in dry soils showed even stronger correlations with CH4 and
CO2, except for CH4 released in the presence of glucose, where correlations with PSD were strongest. Most
soil properties were associated with the first principal component (PC1), which explained 58.1% of the
qualitative differences between the compared soils. The results stressed the significance of the inherent soil
properties in determining the persistence of microorganisms responsible for CH4 and CO2 production over
long storage in air-dry conditions. In fact, all analysed soil properties are related to each other and create
specific habitats which allow microorganisms to persist in unfavourable conditions. Anaerobic
incubations without C amendment resulted in CH4 production in all tested soils, while in some glucose
enriched sand or sandy loam soils methanogens were outcompeted by other microorganisms.
Keywords: soils, methanogens, persistence in dry soils, soil properties, particle size distribution
1. INTRODUCTION
Methane (CH4) is the second most important
greenhouse gas in the atmosphere after carbon
dioxide (CO2). Despite relatively low atmospheric
concentration of CH4 (1.782 ppm in 2006) and short
atmospheric lifetime, it accounts for up to 20-30% of
the global warming effect. Its global warming
potential is up to 40 times higher than CO2, mainly
due to much higher efficiency in trapping radiation
in the atmosphere (Shindell et al., 2009; Owens &
Xu 2011), therefore it is crucial to understand the
processes and factors affecting CH4 fluxes. About
80% of the total CH4 emitted to the atmosphere is
microbially produced during methanogenesis, by a
specific group of methanogens Archaea, in the
absence of free oxygen though anaerobic
decomposition of organic matter. CH4 is formed in
the last step of fermentation where methanogenic
Archaea consume products of activity of the larger
bacterial community including hydrolytic,
fermenting, syntrophic and acetogenic bacteria (Le
Mer & Roger 2001).
The ecology of methanogenic archaea is
complex and not all aspects of soil methanogenic
activity have been elucidated (Inubushi et al., 2003;
Hatano & Lipiec 2004; Xu & Inubushi, 2009; Angel et
al., 2011; Watanabe et al., 2011; Brzezińska et al.,
2012). Methanogenesis is traditionally regarded to
occur only in highly reduced, anoxic environments
such as wetland, rice field soils, mud colcanoes, and
landfills (Megonigal et al., 2004; Frunzeti et al., 2012;
Watanabe et al., 2011). However, low quantities of
CH4 are also produced under unflooded conditions in
various soils (Xu & Inubushi 2009; Angel et al., 2011;
Watanabe et al., 2011). CH4 consumption is conducted
by methanotrophs, the aerobic Proteobacteria.
Methanotrophy occurs under oxic conditions or at the
oxic/anoxic boundaries of soils or sediments (Conrad
2002) where CH4 arisen from methanogenesis or
atmospheric CH4 is consumed. Under experimental
conditions the same soil may produce or consume CH4,
depending on the soil air-water and oxygen status
present (Brzezińska et al., 2012).
CH4 fluxes are known to respond quickly to
seasonal soil moisture fluctuations and changes in
the groundwater level when soil conditions change
from oxic to anoxic (Gliński et al., 2011). Less is
known about the CH4 flux response to more extreme
changes and longer term dry oxic conditions or
flooding for typically aerated soils. According to
current climatic predictions, prolonged dry spells
and altered precipitation patterns causing flooding
will become a very common feature of the temperate
climate (Shindell et al., 2009; Owens & Xu 2011).
The effect of drying and rewetting will become even
more pronounced if soils become water-repellent
and the water infiltration into the soil becomes
restricted (Urbanek et al., 2007). The evermore
extreme oscillations of soil moisture predicted are
likely to alter the survival and activity of various
microorganisms (De Nobili et al., 2006).
Drying and rewetting of the soil is known to
change the soil status and influence the size and
activity of soil microbial populations (Clark &
Hirsch 2008; Chowdhury et al., 2011; Kim et al.,
2012), including methanogenic archaea (Conrad
2002). In environments with regularly occurring
prolonged dry conditions e.g. due to scarce
precipitation, native soil microbial populations
evolve to survive and reactivate in the rainy seasons
(De Nobili et al., 2006). Strict anaerobic bacteria
have the ability to survive under oxic conditions.
Lin et al., (2004) observed that the anaerobic
Geobacter sulfurreducens present in the anoxic
interior zones of soil aggregates, in otherwise oxic
soils, are periodically exposed to oxygen and are
able to tolerate it but become the predominant
microorganism once oxygen becomes limited. The
methanogens also have the ability to endure
desiccation, exposure to high levels of oxygen and
can survive for long periods of time in largely dry
and oxic soils (Liu et al., 2008) despite the fact that
they do not form spores or other resting stages
(Megonigal et al., 2004). Next generation molecular
techniques used in the recent study of Angel et al.,
(2012) revealed that methanogenic microorganism
are not only able to survive short-term oxic
conditions, but are actually globally ubiquitous in
aerated soils and become active once put under wet
anoxic conditions. The discovery of Angel et al.,
(2012) about the abundance of methanogens in
aerated soils raises the question whether soils with
certain specific soil properties have a higher
methanogenic potential than the others.
In the current study we hypothesise that the soil
texture maybe strongly affecting the methanogenic
potential of the soil. Specific particle size distribution
(PSD) in soils creates unique physical conditions
which affects soil structure, water relations, redox
reactions etc., and therefore acts as soil microhabitats
for diverse kinds of microbial biota. It is therefore
very likely that the microbial community structure
and consequently the methanogenic potential strongly
depend on the PSD (Li et al., 2007; Zhang et al.,
2007). In order to test the hypothesis, the
methanogenic potential of soils with different origin
and particle size distribution archived in dried
conditions for over 20 years have been tested. The
archived soils exhibit a very specific habitat for soil
microorganisms due to the lack of external organic
matter addition to the soil. Such isolation from water,
external sources of microbial life and SOM should
make the microbial activity in dry-stored soils more
depended on the inherent properties of the soil rather
than the environmental conditions.
2. MATERIALS AND METHODS
2.1. Soils Description
Sixteen mineral soils, which represent the
main soil types in Poland, were selected for the
study (Bieganowski et al., 2013). The samples were
collected from the topsoil of various agricultural
sites in Poland, air-dried without sieving, and stored
for 23 years at room temperature (about 20oC). Prior
to further incubation trials, the air-dry soils were
sieved through a 2 mm mesh and basic soil
properties were determined. The soils represent a
wide range of textures, with the contents of sand, silt
and clay within 11.3–90.8%, 8.67–76.0%, and 0.52–
12.7%, respectively, the Corg from 0.49 to 2.64%,
and pH (in KCl) from 4.14 to 6.87 (Tables 1-2)
Table 1. Basic characteristics of the tested soils
Soil
No.
Soil typea
pH
KCl
Moist.b
(% w/w)
Corg
(%)
1
Eutric Cambisol
4.44
0.65
0.67
2
Eutric Cambisol
6.19
0.58
0.49
3
Eutric Cambisol
4.14
1.06
0.97
4
Eutric Cambisol
4.45
0.83
0.60
5
Eutric Cambisol
4.85
1.08
0.94
6
Eutric Cambisol
5.09
1.13
1.43
7
Eutric Cambisol
4.64
1.30
1.41
8
Eutric Cambisol
4.45
1.18
1.16
9
Eutric Cambisol
6.17
1.28
0.77
10
Dystric Fluvisol
6.57
1.38
1.31
11
Dystric Fluvisol
5.34
1.76
1.05
12
Dystric Fluvisol
4.51
1.40
1.28
13
Haplic Phaeozem
6.87
1.48
1.06
14
Haplic Podzol
5.57
1.37
0.64
15
Haplic Podzol
6.54
1.07
0.86
16
Mollic Gleysol
5.34
3.07
2.64
a soil type according to FAO; b Moist. – moisture of
air-dry soils before flooding
2.2. The Incubation Experiment
To measure the methanogenic potential of the
soils, 10g samples of soil material were prepared in
60 cm3 glass vials (three replicates per soil and per
treatment), flooded with 10 ml of deionized water or
sterilized glucose solution, sealed with septa and
metal caps, and exposed to two treatments:
1) The N2-variant. Soil was flooded with
water and the headspace was flushed with N2 for
three minutes to remove O2. This treatment
represents the anaerobic conditions without
amendments of the carbon source, which in natural
conditions occurs with a sudden flooding of the soil
or extensive rainstorm with restricted gas exchange
between soil and the atmosphere.
2) The Glucose-variant. Soil was flooded with
glucose solution (5 mg glucose per gram of soil) and
the atmospheric O2 concentration was maintained in
the headspace. The treatment represents soil areas
with so called ‘hot spots’ which form around the
easily available organic matter (e.g. decaying roots).
The addition of easy available C source is expected
to accelerate oxygen consumption, necessary for the
activation of the methanogens.
The soils were incubated statically at 25°C in
the dark for 132 days with frequent measurements of
the concentration of the gases (CH4, CO2, O2).
Table 2. Texture classes and particle size distribution of tested soils - percentage of main fractions and subfractions (in mm)
Soil
No.
Soil texture
classa
Sand
2.0 – 0.05
Sand subfractions
Silt
0.05 – 0.002
Silt subfractions
Clay
< 0.002
2.0
– 1.0
1.0
– 0.5
0.50
– 0.25
0.25
–0.10
0.10
– 0.05
0.05
– 0.02
0.020
– 0.002
1
sand
87.7
15.3
33.8
26.8
8.4
3.4
11.3
5.1
6.2
1.0
2
sand
90.8
4.5
19.6
33.9
26.4
6.5
8.7
4.8
3.9
0.5
3
sand
87.0
1.1
18.9
39.2
22.6
5.2
11.9
6.0
5.9
1.0
4
sandy loam
50.8
3.9
14.1
16.1
9.3
7.4
44.7
16.8
27.9
4.5
5
silt loam
20.4
1.4
2.0
1.71
2.6
12.6
72.2
32.8
39.4
7.4
6
silt loam
14.1
0.0
0.0
0.0
0.9
13.2
75.6
36.7
38.9
10.3
7
silt loam
21.2
2.6
2.5
2.2
3.2
10.7
69.7
30.5
39.2
9.1
8
silt loam
11.3
0.0
0.0
0.0
1.6
9.6
76.0
26.9
49.1
12.7
9
silt loam
18.1
0.0
0.0
0.0
1.0
17.1
76.0
43.5
32.5
5.9
10
sandy loam
55.7
1.4
14.4
18.8
12.6
8.5
39.4
14.4
25.1
4.8
11
silt loam
17.5
0.7
1.1
2.2
4.3
9.2
74.8
31.2
43.7
7.7
12
silt loam
25.3
0.0
0.7
4.1
5.8
14.8
65.2
28.0
37.1
9.5
13
silt loam
19.5
0.0
0.0
0.0
0.6
18.9
73.2
43.5
29.7
7.2
14
sandy loam
69.6
3.7
11.9
20.3
23.6
10.0
26.7
9.2
17.6
3.7
15
silt loam
29.0
2.3
4.8
8.1
5.6
8.2
65.1
31.2
33.9
5.9
16
silt loam
34.0
0.6
3.0
7.6
12.7
10.1
58.6
18.2
40.5
7.3
a soil texture class according to USDA classification
2.3. Methods for determination of gas
concentration, PSD and other soil properties
The concentrations of CH4, CO2 and O2 in the
headspace were measured periodically with a
Shimadzu GC-2014 (Japan) gas chromatograph
equipped with a flame ionization detector (FID) for
CH4 measurements and a thermal conductivity
detector (TCD) for CO2 and O2. The detectors
responses were calibrated using certified gas standards
(Air Products) containing 20.9% O2 in N2, 10 ppm
CH4 and 1% CO2, or 4% CH4 and 10% CO2 in He.
Particle size distribution (PSD) was determined
using the Mastersizer 2000 (Malvern, UK) with a
laser diffractometer within the size range of 0.02 μm
to 2 mm (Ryżak & Bieganowski 2010). The laser
diffraction method involves measuring the intensity
of laser light scattered on the analysed particles.
Hydro G dispersion units were used with the pump
speed set at 1750 rpm and the stirrer at 700 rpm. The
soils were dispersed using ultrasound at 35W for 4
min. In case the obscuration exceeded 10–20%, it
was lowered using the method described earlier. The
intensity of laser light registered on the particular
detectors of the measurement system was converted
to particle size distribution according to the Mie
theory, assuming the following values of the indices:
refraction index 1.52 and absorption index 0.1 for the
dispersed phase, and refraction index of 1.33 for
water as the dispersing phase (Ryżak & Bieganowski
2010; Sochan et al., 2012). Along with the sand, silt
and clay fractions (2–0.050, 0.050–0.002 and < 0.002
mm, respectively), the percentage of subfractions
were determined (Table 2). The textural triangle of
the USDA classification scheme was used to
determine the texture classes of tested soils.
The moisture retained in the air-dry soils was
determined gravimetrically (24 hours at 105oC). The
Corg was determined by TOC-VCPH analyser
(Shimadzu, Japan), and soil pH was measured in 1M
KCl (1:2.5 w/w) after 24 h stabilization at room
temperature. All measurements were conducted in
triplicate and the results were expressed on an oven-
dry weight basis (105°C, 24 h).
2.4. Data Processing and Statistical Analysis
The total cumulative methane released over
132 days was used as a measure of the soil
methanogenic potential. Gases were expressed in
mg CH4-C kg–1 and mg CO2-C kg–1 dry soil; the O2
concentration in the headspace was presented as %
(v/v). The concentrations of the gases were
corrected for solubility in water by using published
values of the Bunsen absorption coefficient
(Glinski et al., 2011). The reading for the α
coefficient was made for the temperature of 25oC,
giving values of 0.029 and 0.829 for CH4 and CO2
respectively. The gas densities of 0.657 and 1.811
mg cm-3 for CH4 and CO2 respectively, were used
for calculation of the gas masses.
The statistical analysis was performed using
Statgraphics Centurion XVI and STATISTICA. An
analysis of variance (Fisher's LSD procedure) was
used to indicate the effect of soil texture classes
(sand, silt loam and sandy loam soils) and soil
conditions established during the incubations (in
the N2- or Glucose- variants) on CH4 and CO2
production. This test also provided information on
the significant factors affecting the methanogenic
potential for examined soils. Simple regression
procedures were performed to describe the
correlations between the produced gases (CH4 and
CO2) and the PSD, Corg, pH and moisture content
retained in the soils. Due to the significant
differences in gas concentrations for both variants,
which could otherwise have been misleading for
the interpretation of the results, only regression
analyses conducted separately for each variant have
been presented in the result.
The principal component analysis (PCA)
was applied to determine the main trends in the
data and the extent of differentiation of CH4 and
CO2 production with regard to different soil
properties. PCA allows the reduction of the
dimensionality of a large number of potentially
correlated variables with the least loss of
information. The PCs were calculated based on
the correlation coefficients matrix, with
eigenvalues greater than 1 being extracted. In the
PCA analysis, the following factors have been
included: CH4, CO2, pH, Corg, moisture retained in
archived soils, and six major PSD fractions
instead of all subfractions. The selection was
based on the results of the regression analyses
(shown in Table 4) and only the fractions with the
highest correlation coefficient were included.
3. RESULTS
3.1. CH4 and CO2 production during soil
incubation
All soils incubated with N2 in the headspace
and without any C amendments (N2-variant) express
some level of methanogenic potential (Fig. 1a).
A lag period before the onset of CH4 production
ranged from 14 days in Eutric Cambisol soils No. 6
and 9, to 121 days in Eutric Cambisol No. 4.. Most
soils start to CH4 release after 20–30 days of
incubation. CO2 evolution from all samples starts
nearly immediately after incubation (Fig. 1b).
The highest microbial activity can be
observed in Mollic Gleysol No. 16 which, over the
whole incubation period, evolved 594.0 ± 18.5 mg
CH4-C kg–1 and 637.2 ± 48.1 mg CO2-C kg–1. The
lowest activity is shown by Eutric Cambisol No. 4
with 0.427 ± 0.33 mg C kg–1 and 87.6 ± 5.86 mg C
kg–1 of CH4 and CO2 produced respectively. Eutric
Cambisol No. 2 showed an unexpectedly low
production of CO2 (69.6 ± 1.66 mg C kg–1), which is
even lower than the amount of CH4 (72.4 ± 0.96 mg
C kg–1).
In most soils incubated in Glucose-variant,
CH4 production starts after a 21 to 90–day lag period
with the total CH4 ranging from 400 to 863 mg CH4-
C kg–1 (Fig. 1c). Despite the fact that some soils start
CH4 production more rapidly (e.g. Eutric Cambisol
No.6 and 9) the overall cumulative CH4 production
was not much different from other soils which show
much longer lag period (i.e. > 60 days in Haplic
Podzol No.14 and Eutric Cambisol No. 2).
Figure 1. Cumulative CH4 (a, c) and CO2 (b, d) over the incubation of 16 archived soils in N2-variant (left graphs)
and in Glucose-variant (right graphs). Inset in graph (d) - changes of O2 concentration in the soil headspace during
incubation with glucose (means ± SD, n = 3)\
However, only on the last incubation day can small
amounts of CH4 be observed in the Eutric Cambisol
No. 4 (5.925 mg CH4-C kg–1), whereas no CH4 is
detected in two Eutric Cambisols (No. 1 and No. 3)
until the end of the incubation. A rapid increase in
CO2 in the headspace is observed in all tested soils
very soon after the start of the incubation (Fig. 1d).
The total released CO2 lies in relatively narrow
range for all soils (597 to 903 mg CO2-C kg–1).
Oxygen concentration for most soils became low
within 14–28 days of incubation. In two soils, O2
rapidly decreases below 1% (v/v) within 5–7 days
(Mollic Gleysol No.16 and Dystric Fluvisol No. 10),
while in soils No. 3 and 4 it remains above 1% (v/v)
for 77 days..All samples (except of soils No. 1 and
3) clearly show an increase in CH4 production once
the O2 levels become very low.
3.2. The effects of incubation conditions
and soil texture on the methanogenic potential of
archived soils
The methanogenic potential of archived soils
presented as the mean CH4 value from each
incubation variant (N2-, Glucose-) and texture class
(sand, sandy loam and silt loam) are presented in
Table 3. Both CH4 and CO2 concentrations are 2.8
times higher (statistically significant) in the
Glucose-variant compared to N2-variant.
As indicated by the F ratio, the effect of
glucose addition is much lower for CH4 than for CO2
concentration (F = 52.4 vs. F = 332.4).
The silt loam soils show the highest
methanogenic potential, which is approximately 5 and
2.5 times higher than in sand and sandy loam soils
respectively (P <0.001). In turn, the average CO2
production from silt loam soils is not significantly
higher than from both sandy loam and sand soils.
Table 3. Mean CH4 and CO2 values (± standard error) accumulated during 132-d incubations of soils flooded after long
storage. Two factor ANOVA (Fisher's LSD procedure) with F-ratios and P-values for both factors: experimental variant
determining incubation conditions and soil texture class (total n=96). Values within a column followed by the same letter
(for a given factor) do not differ significantly at P < 0.05.
Factor
CH4 (mg C kg-1)
CO2 (mg C kg-1)
Experimental variant
F=52.4; P < 0.001
F=332.4; P < 0.001
N2 -variant
174.2 ± 22.5 b
260.9 ± 22.8 b
Glucose -variant
489.9 ± 37.3 a
731.2 ± 12.0 a
Soil texture
F=24.1; P < 0.001
n.s.
Sand
93.0 ± 40.2 b
406.2 ± 74.5 a
Sandy loam
179.1 ± 54.8 b
423.5 ± 64.7 a
Silt loam
449.7 ± 29.1 a
544.8 ± 31.2 a
3.3. The correlations between soil properties
and CH4 or CO2 release
The correlations between the methanogenic
potential and PSD, pH, Corg or moisture retention
after storage are presented separately for two
experimental variants in Table 4. For PSD, the
direction of the relationships depends on the size of
the particles in a given fraction, and is positive for
particles smaller than 0.10 mm, while negative for
particles larger than 0.10 mm. Comparing the
correlation coefficients (r) within particular
incubation variants, the highest positive coefficients
are observed for clay and fine silt fractions.
Table 4. Comparison of the correlation coefficients (r) obtained for CH4 or CO2 released vs. Corg, moisture and PSD
subsfractions (the ranges of subfractions given in mm) in soils incubated for 132 days in two experimental variants
(n = 48 for each variant). The data in bold represent the highest positive and the highest negative correlation coefficients
within a given column.
Soil characteristics
N2-variant
Glucose-variant
CH4
CO2
CH4
CO2
Corg
0.802***
0.839***
0.531***
0.712***
pH (KCl)
n.s.
n.s.
0.341*
-0.595***
Moisture retention
0.729***
0.744***
0.589***
0.582***
PSD subfractions:
mm
Sand
very coarse
2.0-1.0
-0.364*
-0.424**
-0.585***
n.s.
coarse
1.0-0.5
-0.589***
-0.586***
-0.795***
n.s.
medium
0.50-0.25
-0.556***
-0.561***
-0.704***
n.s.
fine
0.25-0.10
-0.319*
-0.347*
-0.349*
n.s.
very fine
0.10-0.05
0.285*
n.s.
0.590***
n.s.
Silt
coarse
0.05-0.02
0.341*
0.304*
0.571***
n.s.
fine
0.02-0.002
0.654***
0.710***
0.694***
0.416**
Clay
< 0.002
0.644***
0.720***
0.692***
0.457**
*,**, *** – significant at P < 0.05, 0.01, and 0.001, respectively; n.s. – not significant;
The coefficient values range from r = 0.416 to
r = 0.720 and have higher values for fine silt in the
case of CH4 and for clay in the case of CO2. In turn,
the strongest negative correlations (except for CO2
in Glucose-variant) are found for CH4 and CO2 in
the coarse sand subfraction (in the range from r = –
0.586 to r = –0.795).
In the N2-variant, the correlation coefficients
with the PSD are slightly higher for CO2 than for
CH4. In the Glucose-variant, positive correlation
between CO2 production and clay or fine silt are
found, but in case of other particles size fractions no
significant correlation can be detected. Only the very
fine sand subfraction (0.10–0.05 mm) is positively
correlated with CH4 production, while all larger sand
subfractions (> 0.10 mm) are negatively correlated
(negative r values). This very fine sand subfraction
is, however, included in some texture classifications
as a silt fraction.
A strong positive correlation (P < 0.001) is
detected between both CH4 and CO2 production and
Corg content, as well as the gravimetric water content
in air-dry soil samples (Table 4). The r values are
even slightly higher for samples incubated in the N2-
variant, than in the Glucose-variant. Correlation with
soil pH exists only for the CO2 production in the
Glucose-variant.
3.4. Results of PCA analysis
The PCA generated three principal
components (PCs) that account for 88.3% of the
underlying variability. The first principal component
(PC1) with an eigenvalue of 7.56 explains 58.1% of
the variance and has a high positive loading for fine
silt, clay, Corg, moisture content, CO2 produced in
the N2-variant and CH4 produced in both variants
(Fig. 2a; Table 5).
Figure 2. The results of principal component analysis: PCA loadings (a) and scores (b) of all tested soils.
High negative loadings are obtained for coarse
and medium sand subfractions. The second principal
component, PC2 (eigenvalue of 2.56) explains 19.7%
of the variance and has a negative loading for CO2
produced in the Glucose-variant. The principal
subsequent component PC3 (eigenvalue of 1.36)
explains much less of total variability (10.5%), and
shows a positive correlation with pH (Table 5). Figure
2b illustrates scores for all tested soils. Positive PC1
scores are dominated by silt loam soils (e.g. Mollic
Gleysol No. 16, Eutric Cambisols No. 6 and 8),
whereas sandy soils (e.g. Cambisols No. 1 and 3) have
high negative PC1 scores. The Mollic Gleysol No. 16
is also characterized by high negative PC2 scores.
4. DISCUSSION
Since soil microbial populations have been
recently highlighted as a major players in the
regulating of major greenhouse gas emissions (e.g.
CO2 and CH4) (Owens & Xu 2011), it is important
to know how they will react to environmental
changes forecast by climatic predictions like
intensive drying or flooding.
Soil air-water conditions strongly influence
the physicochemical status of soil and regulate the
size and activity of soil microbial populations
(Stępniewski & Stępniewska, 2009; Brzezińska et
al., 2011a; Brzezińska et al., 2011b; Gliński et al.,
2011; Włodarczyk et al., 2011, The wetting of dry
soils represents an abrupt step change in soil
biophysical conditions, with critical implications for
biogeochemical cycling; it increases the availability
of soil water, rehydrates microbial cells, increases
microbial metabolism, and mobilizes nutrients (Kim
et al., 2012). De Nobili et al. (2006) reported on the
reactivation of aerobic microorganism’s respiration
in soils stored in air-dry conditions for up to 103
years, while Clark & Hirsch (2008) demonstrated
survival of microbial DNA in dry storage for more
than 150 years. Our study has shown that
methanogenic microorganisms are able to become
active in soils after 20 years of dry storage. The
methanogenic activity can be restored within a few
days or weeks following flooding.
Table 5. Loadings for each variable along PC1, PC2 and PC3 resulting from principal components analysis.
Variable
PC1
PC2
PC3
CH4 (N2-variant)
0.843
-0.360
0.145
CO2 (N2-variant)
0.868
-0.455
0.001
CH4 (Glucose-variant)
0.844
0.173
0.295
CO2 (Glucose-variant)
0.551
-0.754
-0.231
Corg
0.707
-0.531
0.342
pH (KCl)
0.027
0.622
0.734
Moisture retained in dry soils
0.705
-0.372
0.482
Very coarse sand (2.0-1.0 mm)
-0.680
-0.344
-0.354
Coarse sand (1.0-0.5 mm)
-0.892
-0.398
-0.262
Medium sand (0.50-0.25 mm)
-0.850
-0.370
-0.188
Coarse silt (0.05-0.02 mm)
0.690
0.601
0.243
Fine silt (0.02-0.002 mm)
0.911
0.165
0.056
Clay (< 0.002 mm)
0.890
0.136
-0.139
PC1, PC2 and PC3 – first, second and third principal component, respectively; The
highest contribution of each variable is highlighted in bold characters
As can be expected, methanogenesis did not
occur immediately after exposure to anoxic
conditions, but a certain amount of time was
necessary for the microorganisms to respond. In
most soils, CO2 was present in the headspace nearly
immediately following incubation, while a lag
phase ranging from 14 to 132 days was observed
for CH4 production. It should be noted that in other
experiments reported in the literature, such a lag
period (14 days) has also been observed for soils
tested directly after sampling, i.e. incubated
without long storage (Angel et al., 2011;
Brzezińska et al., 2012).
The immediate presence of CO2 in the
sample headspace can be assigned to bacterial
respiration. Additionally, the CO2 presence at the
beginning of the incubation can be partly assigned
to the removal of gases accumulated in soil pores
during the dry storage time (Chowdhury et al.,
2011). The lag phase in the CH4 detection can be
due to time it takes for oxygen and potentially other
alternative electron acceptors to be depleted, as
well as the recovery and growth of the
methanogenic population (Angel et al., 2011). It is
also important to remember that the activity of
methanogenic microorganisms is strongly affected
by other microbial populations such as fermenting
bacteria, iron reducers, syntrophic bacteria etc. that
influence the availability of methanogenic
substrates (Conrad 2002). The activity of other
microorganisms may be especially important for
the recovery of methanogenic microorganisms in
soils exposed to dry conditions for longer periods
of time.
4.1. Influence of different incubation
variant on methanogenic potential
The response in CO2 and CH4 production in
our study was largely dependent on the characteristic
of the individual soil, but also the type of incubation
variant (N2- or Glucose-variant) which simulated
different environmental conditions. The quickest and
the highest response in CH4 and CO2 production was
observed in soils incubated with glucose, which
conforms to expectations as organic amendments are
well known for the stimulation of soil redox
transformations (Gliński et al., 2011) and
methanogenesis (Wang et al., 2013). Glucose
addition to the soil provides a highly labile substrate
for microbial respiration and glucose rather than the
native (more recalcitrant) soil organic matter present
in soil was preferentially used as a carbon source.
Stimulation of both CH4 and CO2 production was
about 2.8-fold greater as compared with the N2-
variant. The significance of this effect was much
stronger for CO2, because all tested soils followed
similar trend of the changes, while the CH4 response
to glucose addition was more diverse (Table 3;
Fig. 1). However, as regression analysis revealed,
native Corg strongly determined CH4 and CO2
release, especially in N2-variant, but also in the
Glucose-variant (Table 4). It is often considered that
native soil organic matter dominated by recalcitrant
forms of carbon is less accessible for the
microorganisms (Hooker & Stark 2001). However,
the observations made by De Nobili et al., (2006),
when investigating the microbial activity of soils
stored for 103 years, showed that the degradation
process of SOM during prolonged soil storage can
make soil carbon more bioavailable. The incubation
of soils with N2 in the headspace created anaerobic
conditions at the start of incubation, but soil organic
matter present in the soil (native) acted as the only
carbon source available for the microorganisms. It is
clear that the lack of an easily available carbon
source was one of the main limitations in the growth
of all microorganisms, when compared with the
glucose variant. Though, an important constraint
was also the deficiency of electron acceptors in the
soil as well as the effectiveness of the anaerobic
bacteria having survived the dry period to reproduce
and reactivate.
In general, CH4 production in the incubation
with glucose started later than in the N2-variant,
likely caused by the presence of oxygen at the
beginning of incubation. In most soils, CH4 release
began once the O2 concentration in the sample
headspace became very low (0.5–0.87% v/v).
Methane evolution in flooded soil has been reported
to start when O2 concentration dropped below 2.5%
v/v (Megonigal et al., 2004). In some soils incubated
with glucose however, CH4 production started just
before the end of the incubation or did not even start,
regardless of the low O2 concentration. This can
indicate that more time was needed to enable
methanogenic populations to overcome the
competition with other anaerobes.
4.2. Effect of soil properties on
methanogenic potential
There has been a clear variability in the
methanogenic potential of individual soils incubated
after long storage, which cannot solely be assigned
to the incubation variant. A clear significant
correlation has been found between CH4 and CO2
production and the PSD. In both experimental
conditions, the methanogenic potential was much
higher in silt loam soils than sandy loam and sand
soils (Table 3). Mean values of total CH4 production
in silt loam soil was approximately 5 and 2.5 times
higher than in sand and sandy loam soils
respectively. Meanwhile the mean total CO2
production in silt loam was only 1.3 times higher
than in coarse textured soils. Despite the fact that the
laser diffraction method used for PSD determination
may underestimate the content of the clay fraction, a
very strong positive correlation has been detected
between both CH4 and CO2 production and the
content of the finest particle fractions like clay and
fine silt (Table 4) (Dobrowolski et al., 2012).
Negative correlations were observed between CH4
and CO2 production and the sand subfractions. It is
therefore evident that the methanogenic potential
increases concurrently with fine soil particle content
and that they play an important role in protection of
methanogens over long, dry and oxic soil conditions.
Several mechanisms can account for such
behaviour. Under field conditions, fine textured soils
have poor drainage and therefore are in general more
prone to anaerobiosis and are expected to favour
methanogenesis (Le Mer & Roger 2001). Sessitsch
et al. (2001) found aerobic and strictly anaerobic
bacteria species to be associated with the clay-sized
fraction and only aerobic species to sand-sized
particles. Small-sized fractions contain the most
microbial biomass in different soils as clay-sized
particles have a higher surface area than coarser
particles, thus facilitating bacterial growth as well as
attachment and protection of microorganisms and
extracellular enzymes (Kögel-Knabner et al., 2008).
Fine textured soils also tend to form aggregates
which create very distinct conditions for water and
organic matter storage (Urbanek et al., 2011) and
therefore allow survival of microorganisms during
prolonged storage. Soil texture is also important for
soil organic matter transformations. The association
of SOM with fine particle fractions, silt and clay,
is considered to play an important role in its
protection (Nicolás et al., 2012). Soil
microorganisms are more abundant in smaller
fractions because of the protection offered by
microaggregates (Lagomarsino et al., 2009).
Methanogens may be able to survive in small
anaerobic microsites imbedded in dry soils, or they
may be protected from O2 by reactive soil minerals
(Megonigal et al., 2004). Additionally, fine textured
soils have the ability to retain more water at higher
suctions, than the coarse textured soils, which can
significantly benefit the survival of microorganisms
(Gliński et al., 2011). On the other hand, the clay
fraction is an important source of trace elements,
including heavy metals (Sȋrbu-Rădășanu et al.,
2013). Nevertheless, Liu et al. (2008) reported that a
liquid culture of Methanobacterium formicicum
could remain viable when mixed well with fresh or
sterile soil, but not when cultured without soil, or
with agar slurry. This suggests that indigenous
methanogens localize within soil compartments to
protect themselves from the damage caused by
gradual drying under an oxic atmosphere.
In contrast to our results, Zhang et al. (2007)
observed, for fertilized paddy soils (but not stored as
air-dry), that CH4 was predominantly produced in the
coarser fractions (which may be contributing to the
storage of labile organic carbon in these fractions),
while more species and a higher diversity of bacteria
survived in the clay sized fraction due to the vicinity
between microbes, access to carbon resource outside
the microaggregates and smaller pore size as
protective agent suitable habitats for microbes rather
than high Corg. Fertilizer application caused more
change to the bacterial community in the clay fraction
and greatly increased bacterium and methanogen
activity in coarser fractions, but only a slight effect on
the methanogenic archaeal community in the particle
size fractions was observed (Zhang et al., 2007).
Apparently, long term storage of air-dry soils resulted
in a shift in the distribution of microorganisms among
soil particles and changed the response of
methanogens to soil rewetting.
Next to the textural properties, the Corg of
archived soils correlated closely with the total CO2
and CH4 production (Table 4). Considerably better
correlations were observed between Corg and CO2
production, rather than with CH4, suggesting that the
availability of organic matter stored in archived soils
is more essential for the activity of other
microorganisms than methanogens. The results
directly correspond with other studies which also
report that the intensity of reductive processes in
flooded soils depends on the content and nature of
organic matter, the ability of the microflora to
decompose SOM as well as availability and nature
of electron acceptors (Le Mer & Roger 2001;
Watanabe et al., 2011). Nevertheless with the
exception of CH4 released in the Glucose-variant,
where fine silt and clay contents were more
important, among soil characteristics, Corg showed
the best correlations with CH4 and CO2 released.
A positive correlation was also found between
the residual water content retained in soils (before
incubations) and the total CO2 and CH4 production
(Table 4). Such a relation can be indirectly linked to
the specific textures of each soil, given that coarse
(sand and sandy loam) soils with larger pores have
lower residual water and Corg contents than finer
soils (silt loam), which are able to store higher
amounts of water, even at high soil suction. The
residual water associated with fine soil particles and
organic matter can therefore create zones for the
survival of microorganisms. The PCA analysis well
illustrate the specific properties of two extremes:
less active sand soils (Cambisols No. 1, No. 3
characterized by low Corg, residual moisture, clay
and silt contents) with their low methanogenic
potential were located at negative PC1 score, while
the most active silt loam soil (Mollic Gleysol No.
16, characterized by high residual moisture, Corg, and
relatively high clay and silt contents) – at high
positive PC1 score, at a distinct distance from other
silt loam soils (Fig. 2b).
5. CONCLUSIONS
Based on the current study, which determined
the methanogenic potential of 16 mineral soils stored
in air-dry conditions over 20 years, it can be
concluded that even after prolonged dry oxic
conditions, soils are able to quickly restore microbial
activity and start methane production within few
days or weeks after flooding. Although we can not
present direct evidence that tested soils retained
active populations of methanogenic Archaea, it was
documented that archived soils do possess potential
for CH4 production without any amendment of the
carbon source. Methanogenic potential shows a
strong positive correlation with finer particle size
fractions like fine silt (0.02-0.002 mm) and clay
(< 0.002 mm). The coarse soil fractions like medium
and coarse sand (0.50-0.25 and 1.0-0.5 mm,
respectively), on the other hand, show strong
negative correlations with the methanogenic
potential. Silt loam soils show significantly, and
approximately 5 and 2.5 times, higher CH4
production than soils with coarser textures (sand and
sandy loam soils, respectively).Therefore finer
textured soils are more likely to become a high
source of methane upon flooding compared to
course textured soils.
The production of CO2 on the other hand, is
not affected by the soil texture if labile carbon
source like glucose has been made available in the
soil. Only in the case when the microorganisms
relied on the native carbon source in the soil (N2-
variant) was both CH4 and CO2 production
correlated with particular PSD fractions. However,
in glucose enriched soils, a significant correlation
for CO2 production was obtained only with the finest
but not with coarser fractions, whereas for CH4
formation all PSD fractions were important.
Nevertheless, the Corg and retained moisture showed
even stronger correlations with methanogenic
potential, except for CH4 released in the presence of
glucose (where the correlation coefficients for Corg
and moisture were lower than for PSD fractions). In
fact, all analysed soil properties are related to each
other and all create specific habitats which allow
microorganisms to persist in long unfavourable
conditions. By applying PCA we can clearly
discriminate between sand soils showing low
methanogenic potential, and silt loam soils
expressing high methanogenic potential. Most soil
properties were associated with the first principal
component (PC1), which explained 58.1% of the
qualitative differences between compared soils.
ACKNOWLEDGEMENTS
The paper was partly financed by the Ministry of
Science and Higher Education/National Science Centre,
Poland (NN310043838). E.U. was supported by the Royal
Society Dorothy Hodgkin Fellowship (DH110189).
REFERENCES
Angel, R., Claus, P. & Conrad, R., 2012. Methanogenic
archaea are globally ubiquitous in aerated soils
and become active under wet anoxic conditions.
The ISME Journal, 6, 4, 847-862.
Angel, R., Matthies, D. & Conrad, R., 2011. Activation
of methanogenesis in arid biological soil crusts
despite the presence of oxygen. PLOS one 6, 5,
e20453.
Bieganowski, A., Witkowska-Walczak, B., Gliński, J.,
Sokołowska, Z., Sławiński, C., Brzezińska, M. &
Włodarczyk, T., 2013. Database of Polish arable
mineral soils: a review. International Agrophysic,
27, 3, 335-350.
Brzezińska, M., Nosalewicz, M., Pasztelan, M. &
Włodarczyk, T., 2012. Methane production and
consumption in loess soil at different slope position.
The Scientific World Journal, 2012, 8 pages.
Brzezińska, M., Rafalski, P., Włodarczyk, T., Szarlip,
P. & Brzeziński, K., 2011a. How much oxygen is
needed for acetylene to be consumed in soil?
Journal of Soils and Sediments 11, 7, 1142-1154.
Bzezińska, M., Sokołowska, Z., Alekseeva, T.,
Alekseev, A., Hajnos, M. & Szarlip P., 2011b.
Some characteristics of organic soils irrigated with
municipal wastewater. Land Degradation and
Development, 22, 6, 586-595.
Chowdhury, N., Burns, R.G. & Marschner, P., 2011.
Recovery of soil respiration after drying. Plant and
Soil, 348, 1, 269-279.
Clark, I.M. & Hirsch, P.R., 2008. Survival of bacterial
DNA and culturable bacteria in archived soils
from the Rothamsted Broadbalk experiment. Soil
Biology and Biochemistry, 40, 5, 1090-1102.
Conrad, R., 2002. Control of microbial methane
production in wetland rice fields. Nutrient Cycling
in Agroecosystems, 64, 1-2, 59-69.
De Nobili, M., Contin, M. & Brooks., P.C., 2006.
Microbial biomass dynamics in recently air-dried
and rewetted soils compared to others stored air-
dry for up to 103 years. Soil Biology and
Biochemistry, 38, 9, 2871-2881.
Dobrowolski, R., Bieganowski, A., Mroczek, P. & Ryżak
M., 2012. Role of periglacial processes in epikarst
morphogenesis: a case study from Chełm Chalk
Quarry, Lublin Upland, Eastern Poland. Permafrost
and Periglacial Processes, 23, 4, 251-266.
Frunzeti, N., Baciu, C., Etiope, G. & Pfanz, H., 2012.
Geogenic emission of methane and carbon dioxide
at Beciu mud volcano, (Berca-Arbănaşi
hydrocarbon-bearing structure, Eastern
Carpathians, Romania). Carpathian Journal of
Earth and Environmental Sciences, 7, 3, 159-166.
Gliński, J., Horabik, J. & Lipiec J., (Eds.), 2011.
Encyclopedia of Agrophysics (Springer
Science+Business Media B.V.
Hatano, R. & Lipiec, J., 2004. Effects of land use and
cultural practices on greenhouse gas fluxes in soil.
Acta Agrophysica, Ser. Monographs, 109, 3-51.
Hooker, T.B. & Stark, J.M., 2012. Carbon flow from
plant detritus and soil organic matter to
microbes—linking carbon and nitrogen cycling in
semiarid soils. Soil Science Society of America
Journal, 76, 903-914.
Inubushi, K., Furukawa, Y., Hadi, A., Purnomo, E. &
Tsuruta, H., 2003. Seasonal changes of CO2, CH4
and N2O fluxes in relation to land use change in
tropical peatlands located in coastal area of South
Kalimantan. Chemosphere, 52, 3, 603-608.
Kim, D.G., Vargas, R., Bond-Lamberty, B. &
Turetsky, M.R., 2012. Effects of soil rewetting
and thawing on soil gas fluxes: a review of current
literature and suggestions for future research.
Biogeosciences, 9, 2459-2483.
Kögel-Knabner, I., Ekschmitt, K., Flessa, H.,
Guggenberger, G., Matzner, E., Marschner, B.
& Lützow, M., 2008. An integrative approach of
organic matter stabilization in temperate soils:
Linking chemistry, physics, and biology. Journal of
Plant Nutrition and Soil Science, 171, 1, 5-13.
Lagomarsino, A., Grego, S., Marhan, S., Moscatelli, M.C.
& Kandeler, E., 2009. Soil management modifies
micro-scale abundance and function of soil
microorganisms in a Mediterranean ecosystem.
European Journal of Soil Science, 60, 1. 2-12.
Le Mer, J. & Roger, P., 2001. Production, oxidation,
emission and consumption of methane by soils: A
review. European Journal of Soil Biology, 37, 1,
25-50.
Li, L., Zhang, X., Zhang, P., Zheng, J. & Pan, G.,
2007. Variation of organic carbon and nitrogen in
aggregate size fractions of a paddy soil under
fertilisation practices from Tai Lake Region,
China. Journal of the Science of Food and
Agriculture, 87, 6, 1052-1058.
Lin, W.C., Coppi, M.V. & Lovley, D.R., 2004.
Geobacter sulfurreducens can grow with oxygen as
a terminal electron acceptor. Applied and
Environmental Microbiology, 70, 4, 2525-2528.
Liu, C.T., Miyaki, T., Aono, T. & Oyaizu, H., 2008.
Evaluation of methanogenic strains and their
ability to endure aeration and water stress. Current
Microbiology, 56, 3, 214-218.
Megonigal, J.P., Hines, M.E. & Visscher, P.T., 2004.
Anaerobic Metabolism: Linkages to Trace Gases
and Aerobic Processes. In: Schlesinger, W.H.
(Ed.), Biogeochemistry. Elsevier-Pergamon,
Oxford, UK, pp. 317-424
Nicolás, C., Hernández, T. & García, C., 2012. Organic
amendments as strategy to increase organic matter
in particle-size fractions of a semi-arid soil.
Applied Soil Ecology, 57, 50-58.
Owens, P.N. & Xu, Z.H., 2011. Recent advances and
future directions in soils and sediments research.
Journal of Soils Sediments, 11, 6, 875-888.
Ryżak, M. & Bieganowski, A., 2010. Determination of
particle size distribution of soil using the laser
diffraction - comparison with aerometric method.
International Agrophysics, 24, 2, 177-181.
Sessitsch, A., Weilharter, A., Gerzabek, M.H.,
Kirchmann, H. & Kandeler, E., 2001. Microbial
population structures in soil particle size fractions
of a long-term fertilizer field experiment. Applied
and Environmental Microbiology, 67, 9, 4215-4224.
Shindell, D.T., Faluvegi, G., Koch, D.M., Schmidt,
G.A., Unger, N. & Bauer, S.E., 2009. Improved
attribution of climate forcing to emissions. Science,
326, 5953, 716-718.
Sȋrbu-Rădășanu, D.-S. & Buzgar, N., 2013. The
geochemistry of major and selected trace elements
in soil from northern area of Ia
ș
city (Romania).
Carpathian Journal of Earth and Environmental
Sciences, 8, 4, 63-74.
Sochan, A., Bieganowski, A., Ryżak, M., Dobrowolski, R.
& Bartmiński, P., 2012. Comparison of soil texture
determined by two dispersion units of Mastersizer
2000. International Agrophysics ,26, 1, 99-102.
Stępniewski, W. & Stępniewska, Z., 2009. Selected
oxygen-dependent process–Response to soil
management and tillage. Soil and Tillage
Research, 102, 2, 193-200.
Urbanek, E., Hallett, P., Feeney, D. & Horn, R., 2007.
Water repellency and distribution of hydrophilic and
hydrophobic compounds in soil aggregates from
different tillage systems. Geoderma, 140, 1-2, 147-155.
Urbanek, E., Smucker, A. & Horn, R., 2011. Total and
fresh organic carbon distribution in aggregate size
classes and single aggregate regions using natural
13C/12C tracer. Geoderma, 164, 3-4, 164-171.
Wang, J., Chen, Z., Ma, Y., Sun, L., Xiong, Z., Huang,
Q. & Sheng, Q., 2013. Methane and nitrous oxide
emissions as affected by organic-inorganic mixed
fertilizer from a rice paddy in southeast China.
Journal of Soils and Sediments, 13, 8, 1408-1417.
Watanabe, T., Wang, G., Lee, C.G., Murase, J.,
Asakawa, S. & Kimura, M., 2011. Assimilation of
glucose-derived carbon into methanogenic archaea
in soil under unflooded condition. Applied Soil
Ecology, 48, 2, 201-209.
Włodarczyk, T., Stępniewski, W., Brzezińska, M. &
Majewska, U., 2011. Various textured soil as
nitrous oxide emitter and consumer. International
Agrophysics, 25, 3, 287-297.
Xu, X.K. & Inubushi, K., 2009. Temperature effects on
ethylene and methane production from temperate
forest soils. Chinese Science Bulletin, 54, 8, 1426-
1433.
Zhang, P., Zheng, J., Pan, G., Zhang, X. & Li, L., 2007.
Changes in microbial community structure and
function within particle size fractions of a paddy soil
under different long-term fertilization treatments
from the Tai Lake region, China. Colloids and
Surfaces, B-Biointerfaces, 58, 2, 264-270.
