American Journal of Botany 88(6): 975–979. 2001.

EFFECT OF AQUATIC WEEDS ON METHANE EMISSION

FROM SUBMERGED PADDY SOIL1

KAZUYUKI INUBUSHI,2,3 HOTAKA SUGII,2,4 SHOUJI NISHINO,2,5 AND

EISHO NISHINO2
2
Faculty of Horticulture, Chiba University, Matsudo 648, Chiba 271-8510 Japan

Paddy fields are one of the dominant anthropogenic sources of methane emission to the atmosphere, and the main passageway of
methane from paddy soil is through the rice plant. However, the effect of aquatic weeds on methane emission from rice paddies has
not been properly evaluated yet. Methane emission from weeded pots and unweeded ones with anaerobic paddy soil was measured
throughout the period of rice growth. More than double the amount of methane was emitted from weeded pots compared with unweeded
ones. Peroxidase activity of rice root was not different between weeded and unweeded pots. However, methanogenic bacteria popu-
lations were higher in weeded pots than in unweeded ones, while methane oxidation activity, measured by the propylene oxidation
technique, was higher in unweeded pots than in weeded ones. Methane oxidation activity of roots from three typical aquatic weeds in
paddy fields, Lipocarpha sp., Rotala indica, and Ludwigia epilobioides, was higher than that of rice plants, while lower stems of these
aquatic plants showed similar or lower activity compared with the same areas of rice plants. These results indicate that the role of
aquatic weeds in paddy soil in methane emission should not be overlooked in evaluating mitigation options for reducing methane
emission from paddy fields.

Key words: aquatic weeds; methane emission; methane oxidation; paddy soil; propylene oxidation; rhizosphere.

Paddy fields are one of the dominant anthropogenic sources
of methane to the atmosphere (estimated as 15% of global
methane emission; IPCC, 1994). The main passageway of
methane emission from anaerobic paddy soils to the atmo-
sphere is through the arenchyma of the rice plant (Oryza sa-
tiva) (Cicerone and Shetter, 1981; Inubushi et al., 1989; Nou-
chi, Mariko, and Aoki, 1990). One possible mitigation option
to reduce methane emission from paddy fields is water man-
agement, such as percolation, mid-summer drainage, and in-
termittent drainage (Inubushi, Muramatsu, and Umebayashi,
1992; Wassmann, Papen, and Rennenberg, 1993; Yagi, Tsu-
ruta, and Minami, 1997). This would introduce oxygen into
anaerobic soil, rendering it aerobic, and in so doing reduce the
formation of methane by enhancing methanotrophic (methane
oxidizing) bacteria in soil (Bosse and Frenzel, 1997; Gilbert
and Frenzel, 1998). Organic farming, which involves the use
of organic matter instead of chemical fertilizer and without
herbicide application, has become popular in Japan (AFFRC,
2000). Both drainage and organic farming may enhance the
growth of weeds in paddy fields. However, the effect of aquat-
ic weeds in paddy fields on methane emission has not been
properly evaluated yet. Therefore it is important to investigate
whether such feedback would exacerbate methane emission by
introducing more organic matter as weeds into the soil or mit-
igate methane emission by methane oxidation.
In this paper, methane emission from weeded and unweeded
1
Manuscript received 7 March 2000; revision accepted 14 September 2000.
The authors thank ex Prof. Yasuji Fukuda for botanical identification. This
work was supported by the Ministry of Agriculture, Fishery and Forestry,
Japan, Research Fund on Agroecological Technology for Controlling Factors
of Global Environmental Changes, and Research Institute of Innovative Tech-
nology for The Earth Fund on Excellent Research Project headed by Prof.
Emer. Iwao Watanabe.
2
Author for reprint requests (Tel 181-47-308-8816, Fax 181-47-308-8720,
e-mail inubushi@midori.h.chiba-u.ac.jp).
3
Current address: Yoshitomi Pharmaceutical Industries Ltd., Yoshiki 3224-
3-508, Yamaguchi 753-0811 Japan.
4
Current address: Kimitsu Agroforestry Highschool, Aoyagi 48, Kimitsu,
Chiba, 292-0454 Japan.
pots with paddy soils were compared in order to estimate the
effect of aquatic weeds on methane flux from paddy soils
throughout the period of rice growth. Microbial activities re-
lated to methane production and oxidation in the soil with rice
plants and typical aquatic weeds in paddy soils were also mea-
sured during crop season to examine possible mitigation op-
tions to reduce methane emission from paddy fields.
MATERIALS AND METHODS
Soils and pot experiments—Sandy gley soil was taken from paddy fields
in Kuju-kuri, Chiba prefecture, Central Japan, and 3.5 kg of the moist sieved
(,7.5 mm) soil was transferred into 18 plastic pots (0.02 m2 3 20 cm depth).
Total carbon, total nitrogen, cation exchange capacity (CEC), and pH(H2O)
of the soil samples were quantified by standard soil analytical method (Brem-
ner and Mulvaney, 1982). Mixed chemical fertilizers were applied as basal
and were equivalent to 100 kg N/ha, 25 kg P/ha, and 42 kg K/ha. Chopped
rice straw (,1 cm length) was applied at 4000 kg/ha on 27 May 1997. Thirty-
five-day-old rice seedlings (Oryza sativa, ‘Nihonbare’) were transplanted 2 d
after fertilizer application and soil submergence. A second fertilizer applica-
tion was made as top dressing at the rate of 20 kg N/ha and 42 kg K/ha at
maximum tillering stage on 25 July. All pots were maintained under flooded
conditions by being placed in a large plastic water pool until harvest on 25
September. All weeds were manually removed weekly from nine pots (weeded
pots), while aquatic weeds, mainly Lipocarpha sp., Rotala indica, and Lud-
wigia epilobioides, all typical paddy weeds in Japan, were allowed to grow
with rice plants in the other nine pots (unweeded pots). Height and tiller
numbers of rice plants were measured weekly. Soil Eh (redox potential) was
monitored to see the effect of weeds on the anaerobic status in soil at 5 cm
depth by Pt electrodes and a pH/Eh meter (Toa Electronics, RM-12P, 29-10-
1 Takadanobaba, Shinjuku-ku, Tokyo, Japan). Root, shoot, and the lower parts
of rice stems (below soil surface and above the root system) were separated
after removal from treatment pots, washed thoroughly, and weighed after dry-
ing at 808C for 24 h on 50, 82, and 111 d after transplanting (DAT). Similarly,
corresponding parts of the aquatic weeds were also taken from the unweeded
pots on the same dates and treated in the same way as the rice plants.
Methane emission—During rice growth, methane emission was measured
almost weekly using the closed chamber method (Inubushi et al., 1989). A
cylindrical acrylic chamber (15 cm diameter 3 1 m height) was placed on

975
976 AMERICAN JOURNAL OF BOTANY [Vol. 88

Fig. 1. Temporal changes in methane flux from weeded (SP) and un- Fig. 2. Temporal changes in soil redox potential (Eh) (5 cm) in weeded
weeded plots (SW). Bars indicate SD of the mean. (SP) and unweeded plots (SW). Bars indicate SD of the mean.

exchange capacity (CEC) 5 76.4 mmol/kg and pH(H2O) 5

the pot with rice for 30 min in the morning (between 1000 and 1200). The
methane mixing ratio inside the chamber was measured at 0, 10, 20, and 30
min after placing the chamber. The gas inside the chamber was sampled with
a syringe, and the mixing ratio of methane in it was quantified by injecting
it into an FID-GC (gas chromatograph with frame ionized detector; Shimadzu
GC-7A, Nishinokyou-kuwahara, Nakagyouku, Kyoto, Japan). The emission
rate was calculated by taking into account the increase in the methane mixing
ratio, the volume of the chamber, and the temperature inside the chamber.
Methanogenic bacteria and their activity in soil—When rice plants and
aquatic weeds were collected, soil samples were taken by truncated plastic
syringe to a depth of 1–5 cm from a spot between the rice hill and inner
surface of the pot. In order to avoid exposing soil samples to the air, they
were immediately transferred into glass bottles and homogeneously mixed
with oxygen-free water. Subsamples of soil, equivalent to 10 g wet mass,
were either incubated anaerobically under N2 headspace at 308C in the dark
in a closed flask to measure methanogenic activity (Chidthaisong, Inubushi,
and Watanabe, 1996; Chidthaisong et al., 1996) or further diluted to estimate
populations of methanogenic bacteria by the most probable number method
(Asakawa and Hayano, 1995).
6.36. Methane flux through rice plants in weeded pots peaked
around 30 DAT and was at a level similar to those in un-
weeded pots until 50 DAT (Fig. 1). However, methane flux
increased rapidly to reach the maximum on 78 DAT in weeded
pots, and this was almost three times higher than in unweeded
plots. As a result, more than double the amount of methane
was emitted during the crop season from weeded pots com-
pared with unweeded ones. Soil Eh was significantly lower in
weeded pots than unweeded pots after 40 DAT except 82 DAT
(Fig. 2), indicating that more oxidized conditions were prob-
ably provided by aquatic weeds.
Plant growth and peroxidase activity of rice roots—Rice
plant height and tiller number were not significantly different
between weeded and unweeded pots (Fig. 3). Dry matter mas-
ses of shoots and roots of rice showed no significant difference
between the two treatments, except being higher in unweeded
pots on 82 DAT, in which roots of weeds were present (Fig.

Methane oxidation activities in soil and plant samples—Soil subsamples

(5 g wet mass) or samples of root, stem, and shoots of rice and aquatic weeds
(2 g wet mass) were transferred into 30-mL Erlenmeyer flasks. To measure
methane oxidation activity, the propylene oxidation method was used (Wa-
tanabe et al., 1995). Thirty milliliters of headspace in the flasks was replaced
with pure propylene and methane to give final concentrations of 20 and 10%
v/v, respectively. The flasks were then incubated for 12 h at 308C in the dark.
At 0, 6, and 12 h of incubation, 0.5 mL of headspace was taken to determine
propylene oxide concentration by FID-GC with Tenax TAt (GL Science, 6-
22-1, Nishishinjuku, Tokyo, 163-1130 Japan).

Peroxidase activity of rice plant and weeds—The peroxidase activity of

roots of rice and aquatic weeds was measured according to Futami (1990).
Briefly, the plant part, equivalent to 2 g wet mass, was placed in a 100-mL
flask and incubated with 50 mL of 40 mg/L alpha-naphthylamine solution and
0.1 mol/L sodium phosphate buffer solution mixture (1:1) for 6 h at 308C in
the dark. After the incubation, 2 mL of the solution in the flask was mixed
with 10 mL distilled water, 1 mL of 1% sulfanil acid, and 10 ppm sodium
nitrite solutions. The optical absorption at 510 nm was read and calibrated
with standard alpha-naphthylamine solutions as a blank test. All the results
were expressed on a dry matter (DM) basis and analyzed statistically by
SYSTATt at 5% significance level using a t test for comparison of the treat-
ment means.

RESULTS
Fig. 3. Temporal changes in plant height (a) and tiller number (b) of rice
Methane flux from weeded and unweeded plots—The soil planted in weeded (SP) and unweeded plots (SW). Bars indicate SD of the
had total carbon 5 7.9 g/kg, total nitrogen 5 0.8 g/kg, cation mean.
June 2001] INUBUSHI ET AL.—EFFECT OF AQUATIC WEEDS ON
Fig. 4. Dry mass of shoot (upper) and root (lower) of rice (R) and weed
(W) in weeded (SP) and unweeded plots (SW). Bars indicate SD of the mean.
4). Similarly, the peroxidase activity of rice roots showed no
difference between these pots, although it decreased gradually
from 50 DAT to harvest (Fig. 5).
Methanogenic bacteria and methane production activity in
soil—Populations of methanogenic bacteria in soil was 4–7
times higher in weeded pots than in unweeded ones (Fig. 6).
Methane production activity was also higher in weeded pots
than in unweeded ones on 82 and 111 DAT.
Methane oxidation activity of soil—Methane oxidation ac-
tivity measured by the propylene oxidation technique was
higher in unweeded pots than in weeded ones, particularly
during the later part of the croping season (Fig. 7).
Fig. 5. Peroxidase activity (mg alpha-naphthylamine · g DM21 · h21) of
rice root in weeded (SP) and unweeded plots (SW). Bars indicate SD of the
mean.
CH4 EMISSION 977
Fig. 6. Populations of methanogenic bacteria (upper) and their activity
(lower) in soil of weeded (SP) and unweeded plots (SW). Bars indicate SD
of the mean. DAT 5 days after transplanting, MPN 5 most probable number.
Methane oxidation activity of roots of three aquatic
weeds—Roots of Lipocarpha sp., Rotala indica, and Ludwigia
epilobioides, typical aquatic weeds in paddy fields in central
Japan, showed higher methane oxidation activity than those of
rice plants. The methane oxidation activity was rated in the
order of Ludwigia . Lipocarpha . Rotala . rice (Fig. 8).
On the other hand, lower parts of the stem of rice plants
showed much higher methane oxidation activity than roots.
The other parts of the aquatic plants showed almost the same
methane oxidation activity as the roots of each plant, and this
Fig. 7. Methane oxidation activity in soil of weeded (SP) and unweeded
plots (SW). Bars indicate SD of the mean. PPO 5 Propylene oxide, DM 5
dry matter.
978 AMERICAN JOURNAL
Fig. 8. Potential methane oxidation activity of plant parts in rice (Oryza) and three species of weeds (Rotala, Lipocarpha, and Ludwigia) growing in flooded
paddy soil. Bars indicate SD of the mean. DM 5 dry matter.
was similar or lower as compared with the same parts of rice
plants.
DISCUSSION
The significance of rice plants in methane emission from
paddy soils to the atmosphere has been recognized since Ci-
cerone and Shetter (1981). Inubushi et al. (1989) estimated
that .90 % of methane flux was through rice plants. Nouchi,
Mariko, and Aoki (1990) examined in detail methane passage
through rice plants. A similar movement of methane from soil
to the atmosphere was also reported for other hygrophytes
(Wagatsuma et al., 1992; Frenzel and Rudolph, 1998). During
methane transport from soil to the atmosphere, 20–80% of
methane was oxidized in rhizosphere soil with rice plants
(Conrad and Rothfuss, 1991; Sass et al., 1991). In natural wet-
lands, aquatic plants were also important in methane oxidation.
King (1994) and Calhoun and King (1998) demonstrated
methane oxidation by methanotrophic bacteria in the rhizo-
sphere of aquatic plants. Gilbert and Frenzel (1995), Bosse
and Frenzel (1997), and Gilbert et al. (1998) found similar
phenomena in rice plants.
The present study showed active methane oxidation in rhi-
zosphere and lower stem parts of typical aquatic weeds as well
as rice plants in paddy fields in Japan. Although the mecha-
nism employed by these weeds to reduce methane emission is
still not clear, peroxidase activity of rice roots was not influ-
enced by weeds (Fig. 5). Reduction of methane flux by weeds
was more likely due to enhancing methanotrophic bacteria
(Fig. 7) or repression of methanogenic bacteria in the rhizo-
sphere (Fig. 6). Both contributions to the reduction of methane
emissions should be evaluated. Since the potential methane
oxidation activity was expressed on a dry matter basis (Fig.
8), the product of (dry matter) 3 (average activity) of weed
roots were estimated to be about twice that of rice roots. How-
ever, rice shoots (lower columns and base of stem) also
showed high methane oxidation activity. Therefore, such es-
timation should include these parts of rice plants. Methane
emission from anaerobic soil to the atmosphere could be re-
OF BOTANY [Vol. 88
garded as the difference between methane production and
methane oxidation. Both activities in this study were measured
separately in soil and plant parts under laboratory conditions.
Therefore, in situ activity in the soil–plant system should also
be examined more carefully in estimating methane oxidation.
Weeds did not reduce the growth of rice plants under these
experimental conditions (Fig. 3), even though the dry mass of
weed shoots was ;14–19% of those of rice shoots (Fig. 4).
However, these results should be examined in detail under var-
ious conditions. Food production could be sustained and meth-
ane emissions might be decreased by reducing the application
of agrochemicals, such as herbicides and by maintaining or
increasing organic matter application to soil. Moreover, meth-
ane emission from paddy soil can also be mitigated by water
management, such as mid-season drainage and other options.
The results in this study indicate that the role of aquatic weeds
in paddy soils in relation to methane emission should not be
overlooked in evaluating mitigation options to reducing meth-
ane emission from paddy fields.
LITERATURE CITED
AFFRC (AGRICULTURE, FORESTRY AND FISHERIES RESEARCH COUNCIL, JA-
PAN). 2000. http://www.maff.go.jp.
ASAKAWA, S., AND K. HAYANO. 1995. Populations of methanogenic bacteria
in paddy field soil under double cropping conditions (rice–wheat). Bi-
ology and Fertility of Soils 20: 113–117.
BOSSE, U., AND P. FRENZEL. 1997. Activity and distribution of methane-
oxidizing bacteria in flooded rice soil microcosms and in rice plants
(Oryza sativa). Applied and Environmental Microbiology 63: 1199–
1207.
BREMNER, J. M., AND C. S. MULVANEY. 1982. Nitrogen-total. In Methods
of soil analysis, Part 2, Agronomy monograph 9, 2nd ed., 595–624.
American Society of Agronomy—Soil Science Society of America, Mad-
ison, Wisconsin, USA.
CALHOUN, A., AND G. M. KING. 1998. Characterization of root-associated
methanotrophs from three freshwater macrophytes: Pontederia cordata,
Sparganium eurycarpum, and Sagittaria latifolia. Applied and Environ-
mental Microbiology 64: 1099–1105.
CHIDTHAISONG, A., K. INUBUSHI, Y. MURAMATSU, AND I. WATANABE. 1996.
Production potential and emission of methane in flooded rice soil micro-
June 2001] INUBUSHI ET AL.—EFFECT OF AQUATIC WEEDS ON
cosms after continuous application of straws. Microbes and Environ-
ments 11: 73–78.
———, ———, AND I. WATANABE. 1996. Methanogenic characteristics of
flooded rice soils in response to glucose amendment. Soil Science and
Plant Nutrition 42: 645–649.
CICERONE, R. J., AND J. D. SHETTER. 1981. Sources of atmospheric methane:
measurements in rice paddies and a discussion. Journal of Geophysics
Research 86C: 7203–7209.
CONRAD, R., AND F. ROTHFUSS. 1991. Methane oxidation in the soil surface
layer of a flooded rice field and the effect of ammonium. Biology and
Fertility of Soils 12: 28–32.
FRENZEL, P., AND J. RUDOLPH. 1998. Methane emission from a wetland plant:
the role of CH4 oxidation in Eriophorum. Plant and Soil 202: 27–32.
FUTAMI, K. 1990. Method of peroxidase activity. In Methods of plant nutri-
tion analysis, 50–52, Hakuyu-sha, Tokyo, Japan.
GIBERT, B., B. ASSMUS, A. HARTMANN, AND P. FRENZEL. 1998. In situ lo-
cation of two methanotrophic strains in the rhizosphere of rice plants.
FEMS Microbiology Ecology 25: 117–128.
———, AND P. FRENZEL. 1995. Methanotrophic bacteria in the rhizosphere
of rice microcosms and their effect on porewater methane concentration
and methane emission. Biology and Fertility of Soils 20: 93–100.
———, AND ———. 1998. Rice roots and CH4 oxidation: the activity of
bacteria, their distribution and the microenvironment. Soil Biology and
Biochemistry 30: 1903–1916.
INUBUSHI, K., K. HORI, S. MATSUMOTO, M. UMEBAYASHI, AND H. WADA.
1989. Methane emission from the flooded paddy soil to the atmosphere
through rice plant. Japanese Journal of Soil Science and Plant Nutrition
60: 318–324.
CH4 EMISSION 979
———, Y. MURAMATSU, AND M. UMEBAYASHI. 1992. Influence of perco-
lation on methane emission from paddy soil. Japanese Journal of Soil
Science and Plant Nutrition 63: 184–189.
IPCC (INTERNATIONAL PANNEL FOR CLIMATE CHANGE). 1994. Climate
change 1994. Cambridge University Press, Cambridge, UK.
KING, G. M. 1994. Association of methanotrophs with the roots and rhizomes
of aquatic vegetation. Applied and Environmental Microbiology 60:
3220–3227.
NOUCHI, I., S. MARIKO, AND K. AOKI. 1990. Mechanism of methane transport
from the rhizosphere to the atmosphere through rice plants. Plant Phys-
iology 94: 59–66.
SASS, R., F. M. FISHER, P. A. HARCOMBE, AND F. T. TURNER. 1991. Methane
emission from rice fields as influenced by solar radiation, temperature
and straw incorporation. Global Biogeochemical Cycles 5: 335–350.
WAGATSUMA, T., K. JUJO, K. TAWARAYA, T. SATO, AND A. UEKI. 1992.
Decrease of methane concentration and increase of nitrogen gas concen-
tration in the rhizosphere by hygrophytes. Soil Science and Plant Nutri-
tion 38: 467–476.
WASSMANN, R., H. PAPEN, AND H. RENNENBERG. 1993. Methane emission
from rice paddies and possible mitigation strategies. Chemosphere 26:
201–217.
WATANABE, I., G. TAKADA, T. HASHIMOTO, AND K. INUBUSHI. 1995. Eval-
uation of alternative substrates for determining methane-oxidizing activ-
ities and methanotrophic populations in soils. Biology and Fertility of
Soils 20: 101–106.
YAGI, K., H. TSURUTA, AND K. MINAMI. 1997. Possible options for mitigat-
ing methane emission from rice cultivation. Nutrient Cycling in Agroe-
cosystems 49: 213–220.