Soil Science Society of America Journal 67:1147-1157 (2003)
© 2003 Soil Science Society of America
DIVISION S-3—SOIL BIOLOGY & BIOCHEMISTRY
Soil Moisture between Rice-Growing Seasons Affects Methane Emission, Production, and Oxidation
H. Xu*,a,
Z. C. Caia and
H. Tsurutab
a Lab. of Material Cycling in Pedosphere, Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008, China
b National Institute of Agro-Environmental Sciences, 3-1-1, Kannondai, Tsukuba 305, Japan
* Corresponding author (hxu{at}issas.ac.cn)
 |
ABSTRACT
|
|---|
Methane (CH4) emissions from rice (Oryza sativa L.) fields are believed to contribute to the greenhouse effect. Earlier studies on CH4 emission were mostly focused on the rice-growing season. The objective of this study was to determine the effects of soil moisture during the non-rice growing season on CH4 emission, production, and oxidation within the subsequent rice-growing season. Five moisture levels ranging from air-dryness to flooding were established in pots during the non-rice growing season. The CH4 fluxes from rice soils in the pots were monitored in a closed chamber and dark incubation was performed to determine CH4 production and oxidation potentials. Both CH4 emission and production increased significantly as the soil got wetter except when it was air-dried. The CH4 oxidation potential was also stimulated by the previous higher soil water content, which therefore buffered emission of the gas as its production increased. Soil water content considerably affected the seasonal variation pattern of CH4 flux and soil redox potential (EH). The higher the soil water content, the quicker soil EH declined and the earlier CH4 emission initiated after rice transplantation. Previous soil water content significantly affected soil organic C content before rice transplantation. Within the rice-growing season both the mean CH4 flux and its production rate were significantly correlated with soil organic C content. Thus water-history-induced change of soil organic C content may have affected soil reduction rate, and then CH4 production and emission within the rice-growing season. The results demonstrate how water management between rice crops can regulate CH4 emission, production, and oxidation during the rice-growing season.
Abbreviations: SWHP, soil water-holding capacity
|
INTRODUCTION
|
|---|
METHANE IS PRODUCED MAINLY by microbial activities in extremely anaerobic ecosystems such as natural and cultivated wetlands, sediments, sewage, landfills, and the rumen of herbivorous animals (Bouwman, 1990). Of the wide variety of the sources, rice fields are the major contributor to the increasing atmospheric CH4 concentration, because of the general upward trend of the area under rice production to meet the food demand of the increasing world population (Dlugokencky et al., 1994). Global annual CH4 emissions from rice fields have been estimated to range from 20 to 150 Tg (IPCC, 1992), 25.4 to 54 Tg (IPCC, 1995), with best estimates at 40 to 60 Tg (Neue, 1993). Further information is needed about the factors controlling CH4 production and oxidation in and the emissions from rice fields to gain a more accurate assessment of the source strength of rice fields in terms of CH4 emission.
Proper water management during the rice-growing season regulates CH4 production and emissions from rice fields. Intermittent drainage is the most promising measure to reduce CH4 emissions (Sass et al., 1992; Cai et al., 1994; Yagi et al., 1997). Unfortunately, this water management practice may cause enhanced N2O emissions from rice fields (Xu et al., 1997, Cai et al., 1997). Soil water regime during the non-rice growing season may also have a strong influence on CH4 emission from rice cultivation. Its effect has not been investigated as intensively as water management during the rice-growing season, though. Comparison of CH4 fluxes from a number of rice fields in China showed that CH4 emissions from the rice fields drained between rice crops were much lower than those from the fields flooded all year (Cai, 1997). The effect of antecedent soil water regime on CH4 emission was further demonstrated by pot experiments (Trolldenier, 1995; Xu et al., 2000) and a field experiment (Cai et al., 2000) that were conducted at only two soil water content levels (flooded and drained). Only 8 to 12% of the rice fields in China are flooded year-round and allowed to lay fallow between rice-growing seasons (Cai, 1997). These fields are distributed mainly in southwest portion of the nation. A major portion of the rice fields in China are not flooded year-round, and when not used to grow rice are either planted with upland crops such as winter wheat (Triticum aestivum L.), oil-seed rape, and green manure, or allowed to remain fallow. As might be expected, the water contents of these soils between rice crops can vary considerably depending on (i) precipitation, (ii) cropping status, (iii) the irrigation system used, and (iv) soil properties. Unfortunately, there are no reports available to be used to evaluate the effect of these varying water levels on CH4 emission.
The emission of CH4 from rice field to the atmosphere is the result of the balance between its production and oxidation. Methane is produced by methanogenic bacteria in the anaerobic layer of paddy soils and oxidized by methanotrophic bacteria in the surface layer of submerged paddy soils and in the rice rhizosphere where both O2 and CH4 are available. Differing soil moisture content between rice crops will probably affect the populations and activity of methanogenic and methanothophic bacteria and cause varying CH4 production and oxidation potentials during rice-growing period. Investigating the effect of the previous soil water content on CH4 production, oxidation, and emission will help us better understand the role of water history in controlling CH4 emission from rice cultivation.
Pot and incubation experiments were conducted from October 1999 to October 2000 to investigate the effect of soil moisture content during the non-rice growing season on CH4 emission, production, and oxidation within the following rice-growing period. Two paddy soils were established at five water content levels ranging from air-dryness to flooding. Soil samples were collected before rice transplantation to measure CH4 production and oxidation potentials and determine soil organic C (including plant debris) content. A close-chamber method was used to measure CH4 fluxes from rice soils in pots.
|
MATERIALS AND METHODS
|
|---|
Soils and Treatments
Soils were established at five moisture levels in 30 greenhouse pots on 26 Oct. 1999. The five moisture levels were maintained through 10 June 2000 whereon all pots were flooded and planted to rice and monitored for their CH4 emissions throughout the growth cycle of that rice.
Two types of loam paddy soils were used in this experiment. One (Soil E) was classified as an Epiaquepts (Soil Survey Staff, 1975) and collected from a flat uniform rice field in Anzhen town, Wuxi county, Jiangsu province. The other (H), a Hapludults, was collected from a similar field in Liujiazhan town, Yujiang county, Jiangxi province. These soils were collected at 15 randomly placed locations from each of the fields (top 20-cm layer) of about 1 ha. The characteristics of Soil E were as follows, pH, 5.57; organic C, 20.1 g kg-1; total N, 1.76 g kg-1. While Soil H had characteristics as follows: pH 5.05, organic C 17.9 g kg-1, and total N 1.45 g kg-1. Both soils were air-dried and passed through a 5-mm sieve.
Thirty pots (20 cm in i.d. and 30 cm in height) were filled each with 6 kg of prepared soil on 26 Oct. 1999. All the soils were fallow and established at the following five water contents during non-rice growing period: (Treatment I) air-dryness (5.37% for soil E and 3.89% for soil H), (Treatment II) 25 to 35%, (Treatment III) 50 to 60%, (Treatment IV) 75 to 85% of soil water-holding capacity (SWHC), and (Treatment V) flooding. The soil water contents specified for each group of pots were maintained by periodically compensating with tap water based on weight changes during the fallow period. The experiment was fully randomized with three replicates.
Rice Cultivation and Measurement of Methane Emission
On 10 June 2000 all soils were flooded and incorporated with basal fertilizers consisting of urea, CaH2PO4, and KCl at the rates of 150 kg N ha-1, 240 kg P ha-1, and 70 kg K ha-1. On 12 June 2000 eight seedlings of ‘Wuyugeng 2’, a rice variety commonly used in Jiangsu province were transplanted at their 3- to 4-leaf stage (Yagi et al., 1996) at the rate of two bunches per pot. The soils received two supplementary applications of urea (top dress) on 8 July and 10 August at a rate of 50 kg N ha-1. All the rice soils were continuously flooded with a water layer of at least 2 cm deep during the rice-growing season. The rice was harvested on 5 Oct. 2000.
Beginning on 12 June 2000 CH4 emissions were measured at an interval of 4 to 5 d with the aid of plexiglass chambers (51 by 51 by 100 cm) and specially designed wooden tables. A hole (22 cm in diameter) was made in the middle of the table so that a pot (21 cm in outer diameter) can easily be put on the table. The hole was lined with a silicon rubber loop glued to the table to prevent gas leakage when pressed by the edge of a pot. A tight fit was also achieved by inserting the flange of the chamber into the square water trough (2 cm wide and 1 cm deep) on the table. Air sampling was usually performed at four times (0, 15, 30, and 45 min).
The CH4 flux, F (mg m-2 h-1), was calculated from the measured concentrations inside a chamber as follows.
where 
(kg m-3) is the density of CH4 at the pressure and temperature recorded inside the chamber, H is the height (in meters) from the water surface to the top of the pot, and dC is the increase in CH4 concentration (µL L-1) inside the chamber in a unit of time (dt, h). The dC was determined from the linear regression of a set of the four data points obtained during a measurement period of 45 min.
Measurement of Methane Production Potential
The CH4 production rates were determined under anaerobic incubation. Twenty grams of thoroughly mixed soils (oven dry-mass basis) collected from each of the 30 pots just before soil flooding were transferred to 120-mL Erlenmeyer flasks. Each 20-g soil sample was placed into its own 120-mL Erlenmeyer flask for incubation and enough water was added to bring the total amount of water to 40 mL, creating a slurry with a soil/water ratio of 1:2. The flask was then sealed with a rubber stopper fitted with a silicon septum that allowed the sampling of headspace gas. Incubation was initiated for all soil samples simultaneously by pumping air from the flask and purging with N2 consecutively for five times through double-ended needles connecting vacuum pump and the flasks. Given the small volume of soil suspension (20 g soil in 40 mL water) and headspace (about 80 mL), this intensive pumping of air and flushing of N2 drove all O2 from liquid and gas phases in the flasks and guaranteed strictly anaerobic incubation of soil samples. Soils had been incubated in the dark for 115 d (the same period as rice-growing season) at 28°C (similar to the seasonally averaged rice soil temperature) with periodic sampling (generally once every week) and repurging with N2. Headspace gas was collected immediately after establishing O2–free condition and 1 d later for analysis of CH4 concentration. The CH4 production rate was determined from the daily increase of CH4 concentration and the headspace volume. Before each sampling flasks were shaken vigorously by hand for 30 s to ensure that CH4 entrapped in soil suspension was negligible.
Measurement of Methane Oxidation Potential
Twenty grams of thoroughly mixed soils (oven dry-mass basis) collected from each of the 30 pots just before soil flooding were transferred to 120-mL Erlenmeyer flasks. The flask was then sealed with a rubber stopper fitted with a silicon septum that allowed the injection of CH4 and the sampling of headspace gas. The initial CH4 concentration in headspace was about 8500 µL L-1 by injecting 1 mL of pure CH4. The actual concentration was determined 2 h after pure CH4 was injected to ensure a homogeneous CH4 distribution inside the flask. Then the soils were incubated in the dark at 28°C. The CH4 oxidation potentials were determined by monitoring the change of CH4 concentration in the headspace. Periodically (every 12 h or so) 0.2-mL headspace gas was sampled with a 1-mL syringe for CH4 concentration analysis. After sampling, 0.2 mL of air was injected into the headspace to compensate the pressure reduction. The CH4 concentration change caused by the injection of 0.2 mL of air was adjusted by the change of CH4 concentration in the control that had the same initial CH4 concentration but didn't contain soil. The sampling interval was adjusted to more than 30 h when CH4 concentration of Treatment I was still high and decreased slowly while CH4 concentration of the other four treatments declined to a low and stable level.
The CH4 oxidation potential was assessed under two separate protocols: (i) at the same water contents as those during the non-rice growing season and (ii) at a identical content level (80% of SWHC) by either adding distilled water or air drying the soil. Incubation periods were 175 h for the first and 239 h for the second.
The incubation experiments were also performed in triplicate and the average was used to represent CH4 production and oxidation potentials of soil in the corresponding pot.
Methane Analysis
The CH4 concentrations were quantified on a gas chromatograph (Shimadzu 12 A) equipped with a flame ionization detector (FID) and a 2-m Poropak Q (80/100 mesh) column. The temperature of the oven was 80°C, while those of the injector and the detector were both 200°C. The carrier gas (N2) flow rate was 30 mL min-1. Methane had a retention time of 0.9 min and its peak was integrated on a Shimadzu chromatopac C-R6A integrator. Calibration was performed using local standard gases of which the CH4 concentration was calibrated in National Institute of Agro-environmental Sciences (Kannondai 3-1-1, Tsukuba, Japan).
Measurement of Soil Redox Potential
When CH4 fluxes were monitored, the EH of rice soils were simultaneously measured by using Pt-tipped electrodes (Hirose Rika Co., Ltd. Japan) and an oxidation-reduction potential meter (Toa RM-1K, Toa Co., Ltd. Japan). For the measurements of soil EH, the electrodes were inserted into the soil at a depth of 10 cm and maintained there throughout rice-growing period. The potentials measured against AgCl reference electrode were reported in reference to the H2 electrode. All soil EH measurements were made in triplicates.
Determination of Soil Organic Carbon
Organic C (including plant debris) contents of soils collected from each of the 30 pots just before flooding were analyzed by the dry combustion method (Nelson and Sommers, 1996).
Mean Value Calculation and Statistics
We use weighted averages method to calculate the mean value of CH4 flux, CH4 production, and soil EH on the basis of their individual data and measurement intervals during the observation period.
Statistical analysis of the data was performed on the replicates by ANOVA. If the main effect was significant at P < 0.05, a post hoc separation of means was done by least significant difference (LSD) test. Statistical analysis was conducted with STATISTICA 5.1 for windows (StaSoft Inc., Tulsa, OK). Correlation coefficients were calculated by linear least-square regression analysis.
|
RESULTS
|
|---|
Methane Emission
Pronounced seasonal variations in CH4 fluxes were observed for all the five treatments (Fig. 1)
. The fluxes began to increase 0 to 34 d after flooding and peaked on August 14 (Day 63), then decreased and, for the most part, stabilized toward the end of the rice-growth period. The time when CH4 emission started to occur differed greatly among treatments. The soil in Treatment V was already emitting CH4 at the time rice was transplanted into it. The onset of emissions took longer with the other treatments: about 11 d after transplanting for Treatment I; 16 d for Treatment IV; 24 d for Treatment III, and 34 d for Treatment II. Soil with higher water content during the non-rice growing season showed earlier CH4 emission for all soil moisture levels except air-dryness.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 1. Temporal variations of CH4 fluxes during the rice-growing period. (a) Soil E (Epiaquepts); (b) Soil H (Hapludults); I, II, III, IV, and V represent air-dryness water condition, 25 to 35, 50 to 60, 75 to 85% of soil water-holding capacity and flooding, respectively, during non-rice growing season. Bars indicate ± one standard deviation for Treatments II, III, and V; results with other treatments showed a similar range of standard deviation.
|
|
The difference of soil EH among five treatments was conspicuous only at the early stage of rice growth (Fig. 2)
. A strict anaerobic condition with soil EH as low as -147 (Soil E) and -132 mV (Soil H) was already established when rice was transplanted for Treatment V. In contrast, soil EH of the other four treatments was high (225–444 mV, Soil E; 234–470 mV, Soil H) on the day of rice transplantation and then decreased at various rates. Soil EH of Treatment I dropped most rapidly to below 0 mV within 11 d, followed in order by Treatments IV (16 d for Soil E and 20 d for Soil H), III (34 d for Soil E and 30 d for Soil H), and II (43 d for Soil E and 38 d for Soil H). With the increase of water content during the non-rice growing season, soil reduction after rice transplantation became fast for all cases except air-dry water condition. The high soil EH of Treatments I, II, III, and IV corresponded closely to the negligible CH4 emission during the early stage of rice growth (Fig. 1 and 2).
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 2. Temporal variations of soil EH during rice-growing period. (a) Soil E (Epiaquepts); (b) Soil H (Hapludults); I, II, III, IV, and V represent air-dry water condition, 25 to 35, 50 to 60, 75 to 85% of soil water-holding capacity and flooding, respectively, during non-rice-growing season. Bars indicate ± one standard deviation for Treatments II and V; results with other treatments showed a similar range of standard deviation.
|
|
As shown in Fig. 3
, soil temperatures at soil depths of 0, 5, and 10 cm had a similar temporal variation pattern and fluctuated from 20.9 to 32.3°C during the rice-growing season. The highest soil temperature occurred on August 14 (Day 63) when seasonal peak CH4 fluxes appeared (Fig. 1 and 3). Please note the date 14 August (Day 63) in Fig. 1 and 3. Although soil temperatures after that date were lower than they were before it, CH4 fluxes were instead higher.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 3. Temporal variations of soil temperatures at depths of 0, 5, and 10 cm during rice-growing period.
|
|
The seasonal mean CH4 fluxes and soil EH as well as organic C contents of soils collected before flooding is summarized in Table 1
. The highest mean CH4 flux and soil organic C content were from Treatment V, followed in order by Treatments IV, I, III, and II (CH4 flux) and I, IV, III, and II (C content). In contrast, the mean soil EH of Treatment V was the lowest followed in order by Treatments I, IV, III, and II. These results show that in all except air-dryness, as soil water content increased during the non-rice growing period, the organic C content and CH4 emission increased and the soil EH decreased significantly.
View this table:
[in this window]
[in a new window]
|
Table 1. The weighted mean CH4 flux and soil EH during rice growth season and soil organic C (including plant debris) contents immediately before rice transplantation.
|
|
Figure 4
demonstrates that seasonally averaged CH4 fluxes correlated positively with organic C content of soils collected before soil flooding (data were calculated on water treatment basis), but only when the data from the two tested soils were fitted separately.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 4. Relationship between mean CH4 fluxes during rice-growing season and soil organic C contents before rice transplantation. Data in this figure were calculated on water treatment basis.  Soil H (Hapludults), • Soil E (Epiaquepts).
|
|
The rice yields of different treatments are presented in Table 2 . As shown in Tables 1 and 2, all five of the water regimes we tested produced similar rice yields, though CH4 emissions differed significantly among them.
Methane Production Potential
Temporal variations in CH4 production rates during anaerobic incubation period (corresponding to rice-growing season) are shown in Fig. 5
. Similar to the variation in time of CH4 initiation that we found after rice transplantation (Fig. 1), we also found a variation in time after anaerobic incubation until the beginning of CH4 production. Production of CH4 began immediately after incubation for Treatment V. It took 4 d for production to begin in Treatments I (Soils E and H) and IV (Soil H), 11 d for it to begin in Treatment IV (Soil E), 18 d in Treatment III, and 25 d in Treatment II. The CH4 production rates of all the five treatments reached one striking peak value after 25 (V), 60 (II, III, and IV), and 66 (I) d of incubation, then declined quickly in 1 to 2 wk, and thereafter declined slowly to a low level and became stable.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 5. Temporal variations of CH4 production rates during a 115-d incubation. (a) Soil E (Epiaquepts); (b) Soil H (Hapludults); (c) Treatment V; I, II, III, IV, and V represent air-dry water condition, 25 to 35, 50 to 60, 75 to 85% of soil water-holding capacity and flooding, respectively, during non-rice-growing season. Bars indicate ± one standard deviation for Treatments II, III, IV and V; results of Treatment I showed a similar range of standard deviation as Treatment III.
|
|
Table 3
shows the proportion of CH4 emission and production after Day 63 to the amount summed during the observation period. Differed obviously from CH4 emission, CH4 production became much lower during the late incubation stage.
View this table:
[in this window]
[in a new window]
|
Table 3. Proportion (%) of CH4 emission and production after Day 63 to the amount summed over 114-d (emission) and 115-d (production) observation.
|
|
The mean CH4 production rates of different treatments during the incubation period are listed in Table 4
. Treatment V appeared to have the highest CH4 production rate, followed in sequence by Treatments IV, I, III, and II, which showed that CH4 production potential increased significantly with the increase of the previous soil water content, except air-dryness.
As was the case with the mean CH4 flux during rice-growing season, the mean CH4 production rate was also significantly and positively correlated with soil organic C content (r = 0.79, Soil E; r = 0.76, Soil H).
There was a significant positive correlation between CH4 production and emission (Fig. 6)
. Data in Fig. 6 were calculated on pot basis for both Soils E and H. Differed from Fig. 4 where only the CH4 flux and soil organic C data of the same soil can be fitted by linear correlation, Fig. 6 offered a good linear fit to all the data of both Soils E and H.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 6. Relationship between the mean CH4 fluxes during rice-growing season and production rates during 115-d incubation. Data in this figure were calculated on pot basis.
|
|
Methane Oxidation Potential
Molecular O2 is rapidly depleted after flooding of paddy soils and remains present only in the thin surface layer of submerged soil and possibly also in the rice's rhizosphere. Therefore we refer to the CH4 consumption rate measured in the incubation under the mixture of CH4 and air as CH4 oxidation potential.
The CH4–oxidizing potentials differed markedly among soils with varying water content during both non-rice growing and incubation period (Fig. 7)
. While CH4 concentration of Treatment V decreased sharply to 90 (Soil E) and 157 (Soil H) µL L-1 after 42 h of incubation, it took 103 (IV), 139 (III), and 175 h (II) to reach a similar low level of CH4 concentration of Treatment V. Air-dried soils showed greatly inhibited CH4 oxidation potentials. The CH4 concentration of Treatment I was still as high as 3312 (soil E) and 3667 (soil H) µL L-1 after 175 h of incubation. The CH4 oxidation potentials were stimulated by higher soil water content during non-rice growing and the incubation period.
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 7. Time course of CH4 oxidation in paddy soils pretreated by varying water content and incubated under the same historic water condition. (a) Soil E (Epiaquepts); (b) Soil H (Hapludults); I, II, III, IV, and V represent air-dry water condition, 25 to 35, 50 to 60, 75 to 85% of soil water-holding capacity and flooding, respectively, during non-rice-growing season. Bars indicate ± one standard deviation.
|
|
It is difficult to differentiate the contribution of previous soil water condition to CH4 oxidation potential from that adopted during incubation period (Fig. 7). To emphasize the exclusive effect of soil moisture history on CH4 oxidation capacity, the soils of the five treatments were all incubated under 80% SWHC (Fig. 8)
. The concentration of CH4 dropped quickly to about 100 µL L-1 in 73 h (Treatment V), whereas, 96 (IV), 120 (III), and 138 h (II) were required for CH4 concentrations to reach a similar low level. Being incubated for 239 h, the remaining CH4 concentration of Treatment I was still at a relatively high level (1293 and 1732 µL L-1 for Soil E and H). As soil water content increased during the previous non-rice growing season, CH4 oxidation potential increased accordingly.
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 8. Time course of CH4 oxidation in paddy soils pretreated by varying water content but incubated under an identical water condition (80% of soil water-holding capacity). (a) Soil E (Epiaquepts); (b) Soil H (Hapludults); I, II, III, IV and V represent air-dry water condition, 25 to 35, 50 to 60, 75 to 85% of soil water-holding capacity and flooding, respectively, during non-rice-growing season. Bars indicate ± one standard deviation.
|
|
Compared with incubation at the same water status as during the non-rice growing season (Fig. 7), the CH4 oxidation capacity of the soils when incubated at 80% SWHC was almost the same (Treatments I and IV), higher in varying degrees (Treatments II and III) and lower (Treatment V) (Fig. 8). In contrast to Fig. 7, Fig. 8 displays closer time courses of CH4 oxidation for Treatments II and III.
|
DISCUSSION
|
|---|
Studies into the effect of the soil water regime on CH4 emission have focused mainly on the period of rice growth (Sass et al., 1992; Bronson et al., 1997; Yagi et al., 1997). Few experiments have explored the importance of soil water regime during the antecedent non-rice growing season in controlling CH4 emission from rice cultivation. The only two-pot experiments (Trolldenier, 1995; Xu et al., 2000) and a field experiment (Cai, 1997) were conducted at two soil water content levels (flooded and drained) and gave results of much higher CH4 emissions from rice soils flooded year-round. Our results based on five soil water content levels not only further confirm the importance of the antecedent soil water regime in affecting CH4 emissions, but also suggest its important role in controlling CH4 production and oxidation potentials during the rice-growing season.
Methane generation is the terminal step in anaerobic microbial decomposition of organic matter and thus requires a sufficient low soil EH. The progress of soil reduction is controlled by the relative abundance of electron donors and electron acceptors in the soil (Yagi et al., 1995). In the absence of O2, the main electron-accepting chemicals are NO-3, Mn4+, Fe3+, and SO2-4. The main electron-donor is readily decomposable organic matter (Yagi et al., 1994). The formation of CH4 occurs only after NO-3, Mn4+, Fe3+, and SO2-4 are reduced in thermodynamically sequential order (Neue, 1993). Soil organic matter acts not only as the substrate for CH4 production, it also helps stimulate soil reduction and creates a strict reductive condition for CH4 production. Thus soils containing high percentages of organic matter are likely to show a high production level of CH4. Lower soil water content during the non-rice growing season caused lower content of organic C (Table 1) before rice transplantation. This led to slower soil reduction and delayed CH4 emission after rice transplantation, higher soil EH and lower CH4 flux and production rate averaged over the observation period (Fig. 1 and 2, Tables 1 and 4). This helps explain why there are significant correlations between the seasonally averaged CH4 fluxes and production rates and soil organic C contents before rice transplantation. Other researchers have also demonstrated a good relationship between CH4 production and soil organic C content (Crozier et al., 1995; Wang et al., 1999).
Though CH4 emission was significantly affected by soil organic C content before rice transplantation, only the data of the same soil can be fitted by linear correlation (Fig. 4). Moreover, the change of soil organic C content was at least partially caused by differing water content during the non-rice growing season, indicating the importance of prior soil water status in controlling CH4 emission during rice cultivation.
Besides soil organic matter degradation, methanogens in rice soils can get additional C substrate from root exudation, decay and plant senescence. The seasonal CH4 emission maxima were attributed to not only the highest soil temperature, but also the added C substrate from root exudation at the reproductive stage of rice growth (Fig. 1 and 3). One of the main differences between pot and incubation experiments in this study is with (pot) and without (incubation) rice planting. Though soil temperature and CH4 production rates became lower after Day 63, CH4 fluxes were same or even higher (Fig. 1, 3, and 5; Table 3), which might be because of more substrate supply through root decay and plant senescence in the late vegetative period.
Methane production relies exclusively on methanogenic bacteria, which are active only under strict anaerobic conditions. Asakawa and Hayano (1995) found that methanogenic populations were almost constant in the soils with either summer rice under flooded conditions or winter wheat for 2 yr irrespective of soil moisture regime. However, it is possible that the methanogenic populations may be reduced if the soils are exposed to atmospheric O2 and the reduction may become more evident with the decrease of soil water content during the non-rice growing season because methanogenic bacteria are sensitive to O2. The effect of soil water content during the non-rice growing season on methanogenic populations and activities might be another reason why the soil with higher water content produced and emitted more CH4 during rice-growing season. Though soil of Treatment I had higher organic C content than Treatment IV, CH4 production rates of these two treatments showed a reverse trend (Table 2), which might also be explained by the possible difference of methanogenic populations and activities. Unfortunately we didn't test soil methanogenic populations and activities in this experiment.
The traditional options recommended for reducing CH4 emissions from rice fields have been focused on the rice-growing season. Since the previous soil water regime affects significantly CH4 emission, and the highest CH4 flux was recorded in a rice field flooded year-round (Khalil et al., 1991), attention should also be paid to the non-rice growing season to find more mitigation options. In China, the 8 to 12% of the rice fields that are both flooded and fallow between the rice-growing seasons contribute the dominant amount (60%) of the total CH4 emissions from rice fields (Cai, 1997). If the irrigation and drainage facilities for these rice fields could be improved substantially and the flooded water could be drained completely during the non-rice growing season, the total CH4 emissions from China's rice fields would be significantly reduced. However, this mitigation option seems somewhat impractical due to water shortage in these regions. There would be a danger if the fields were drained after a rice crop and couldn't receive enough water to be entirely reflooded for the next rice crop.
While CH4 emission from rice field is of concern to some, the yield of that rice crop is of at least as important a consideration to others. Although soil moisture regime during the antecedent non-rice growing season greatly affected CH4 emissions from rice soils in pots, it had no significant effect on rice yields (Tables 1 and 2). This indicates that we would not risk rice yield loss when we try to reduce CH4 emission by means of adjusting soil water content during the non-rice growing season.
Soil moisture influences CH4 oxidation in two aspects; (i) diffusion and supply of substrate and O2 and (ii) methanotrophic activity. Numerous studies have found that there is an optimum intermediate soil water content for CH4 oxidation (Bender and Conrad, 1995; Whalen and Reeburgh, 1996; Cai and Yan, 1999). As soil moisture content decreases below the optimum value, CH4 uptake becomes gradually inhibited because of an increase in physiological water stress of methanotrophs although more O2 and CH4 could be supplied through diffusion. Whereas when soil water content increases above the optimum level, CH4 oxidation may be gradually suppressed by the slowing of CH4 diffusion and O2 transport in the soil. One trait of all available literature in this area is that the research reported was conducted on soils with the same water history. Our research was the first attempt to explore CH4 oxidation potentials for soils with varying water backgrounds. Our results show that CH4 oxidation potential increases with the increase of the previous soil water content from air-dryness to flooding (Fig. 7 and 8). Because it had been flooded during the non-rice growing season, the soil of Treatment V had undergone long-term anaerobic conditions, which resulted in reasonably a high CH4 concentration when it was collected for CH4 oxidation experiment. Previous studies have demonstrated that the number of methanotrophic bacteria increases if CH4 concentration is high enough (Bender and Conrad, 1995; Kightley et al., 1995; Arif, 1996). Therefore, it may be the high number of methanotrophic bacteria that causes the highest CH4 oxidation potential for Treatment V though gas diffusion is greatly impeded under saturated water conditions. On the other hand, soil that had been air-dried during the non-rice growing season had the lowest oxidation potential of all soils irrespective of soil moisture condition employed during incubation, which indicates the strong inhibitory effect on CH4 oxidation of long-term extremely low water content.
We have found that not only production but also oxidation potential of CH4 increases with the increase of soil water content during the non-rice growing season. While the highest CH4 production rate is 13.3 times higher than the lowest, the corresponding value for CH4 flux is only 6.1 (Tables 1 and 4). It seems that paddy soils with high CH4–producing potential also show high CH4–oxidizing capability. This self-adjusting of a soil's CH4 oxidation capacity is important to buffer CH4 emission from rice field when CH4 production increases.
The low water content of air-dry soils inhibits activity of soil microorganisms and suppresses decomposition of soil organic matter. Therefore the organic C content of air-dry soil was an exception to the regularity between organic C content before rice transplantation and soil water content during the non-rice growing season. This exception to the regularity, together with the lowest CH4 oxidation capability may be the main reason why CH4 flux, CH4 production rate, soil EH, and their temporal variations of air-dryness treatment didn't follow the regularities mentioned in the corresponding sections of this paper.
While we chose air-dryness as one of the water treatments in this study to achieve a largest possible water-content-change range, we realize that precipitation or irrigation between rice crops makes it nearly impossible for such a condition to continue in a field for a whole season. A soil that received only 327.4 mm of rainfall during the non-rice growing season showed a similar low CH4 flux to that of Treatment II, though this soil reached air-dryness to some depth in drought season (Xu et al., 2000). Therefore, the results of air-dry soils in this study were of more theoretical than practical value and then a positive correlation might actually exist between soil water contents during the non-rice growing season and CH4 emissions from rice fields. This inference has been confirmed by a finding that there was a significant correlation between CH4 fluxes measured in China's rice fields and simulated soil moisture contents during the non-rice growing season (Kang et al., 2002).
Besides CH4, two additional important greenhouse gases, CO2 and N2O, can be produced in and between rice crops. Mineralization of soil organic matter can release CO2, nitrate, and ammonium, the latter two of which can be changed to N2O by either nitrification or denitrification. All the processes related to N2O and CO2 production in soil are influenced by soil water content. Thus, more researches should be done to investigate the effect of soil water content during the non-rice growing season on yearly emissions of all the major greenhouse gases to quantify the overall potential greenhouse effect affected by prior soil moisture status.
|
CONCLUSIONS
|
|---|
Both CH4 production and oxidation potentials were stimulated by higher soil water content during the previous non-rice growing season. As soil water content increased, so too did CH4 production and, to a lesser extent, CH4 emission. The lower level of emission may be due to the increased buffering effect of CH4 oxidation by the soil with higher water content. Soil reductive capacity change caused by differing moisture content might be one of the main reasons why soils with varying water history produced and emitted differing amount of CH4 during the rice-growing season. Since the highest CH4 emission was from soils flooded during the non-rice growing season, special attention should be paid to rice fields flooded throughout the year.
|
ACKNOWLEDGMENTS
|
|---|
The authors express their sincere gratitude to the responsible editors and anonymous reviewers for corrections to and suggestions for the manuscript. This project was supported by the National Key Basic Research Support Foundation of China (G1998011805), Chinese Academy of Sciences (KZCX2-302) and the Red Soil Ecological Experimental Station, Chinese Academy of Sciences.
Received for publication May 7, 2002.
|
REFERENCES
|
|---|
- Arif, M.A.S. 1996. Agricultural factors affecting methane oxidation in arable soil. Biol. Fertil. Soils
21:95–102.
- Asakawa, S., and K. Hayano. 1995. Populations of methanogenic bacteria in paddy field soil under double cropping conditions (rice-wheat). Biol. Fertil. Soils
20:113–117.
- Bender, M., and R. Conrad. 1995. Effect of CH4 concentration and soil condition on the induction of CH4 oxidation activity. Soil Biol. Biochem.
27:1517–1527.
- Bouwman, A.F. 1990. Exchange of greenhouse gases between terrestrial ecosystems and the atmosphere. p. 61–127. In A.F. Bouwman (ed.) Soils and the greenhouse effect. John Wiley and Sons, Chichester, U.K.
- Bronson, K.F., H.-U. Neue, U. Singh, and E.B. Abao, Jr. 1997. Automatic chamber measurements of methane and nitrous oxide flux in a flooded rice soil: I. Residue, nitrogen, and water management. Soil Sci. Soc. Am. J.
61:981–987.[Abstract/Free Full Text]
- Cai, Z.C. 1997. A category for estimate of CH4 emission from rice paddy fields in China. Nutrient. Cycling Agroecosyst.
49:171–179.
- Cai, Z.C., and X.Y. Yan. 1999. Kinetic model for methane oxidation by paddy soil as affected by temperature, moisture and N addition. Soil Biol. Biochem.
31:715–725.
- Cai, Z.C., H. Tsuruta, and K. Minami. 2000. Methane emission from rice fields in China: Measurements and influencing factors. J. Geophys. Res.
105:17231–17242.
- Cai, Z., G. Xing, X. Yang, H. Xu, H. Tsuruta, K. Yagi, and K. Minami. 1997. Methane and nitrous oxide emissions from rice paddy fields as affected by nitrogen fertilizers and water management. Plant Soil
196:7–14.
- Cai, Z.C., H. Xu, H. Zhang, and J. Jin. 1994. Estimate of methane emission from rice paddy fields in Taihu region, China. Pedosphere
4:297–306.
- Crozier, C.R., I. Devai, and R.D. Delaune. 1995. Methane and reduced sulfur gas production by fresh and dried wetland soils. Soil Sci. Soc. Am. J.
59:277–284.[Abstract/Free Full Text]
- Dlugokencky, E.J., L.P. Steele, P.M. Lang, and K.A. Masarie. 1994. The growth rate and distribution of atmospheric methane. J. Geophys. Res.
99:17021–17043.
- IPCC. 1992. Climate change, the supplementary report to the IPCC scientific assessment. Cambridge University Press, Cambridge, UK.
- IPCC. 1995. Climate change, impacts, adaptations and mitigation of climate change: Scientific-technical analyses. Cambridge University Press, Cambridge, UK.
- Kang, G.D., Z.C. Cai, and X.Z. Feng. 2002. Importance of water regime during the non-rice growing period in winter in regional variation of CH4 emissions from rice fields during following rice growing period in China. Nutrient. Cycling Agroecosyst.
64:95–100.
- Khalil, M.A.K., R.A. Rasmussen, M.X. Wang, and L. Ren. 1991. Methane emissions from rice fields in China. Environ. Sci. Technol.
25:979–981.
- Kightley, D., D.B. Nedwell, and M. Cooper. 1995. Capacity for methane oxidation in landfill cover soils measured in laboratory-scale soil microcosms. Appl. Environ. Microbiol.
61:592–601.[Abstract]
- Nelson, D.W., and L.E. Sommers. 1996. Total Carbon, organic carbon, and organic matter. p. 961–1010. In D.L. Sparks et al. (ed.) Methods of soil analysis. Part 3. SSSA, Madison, WI.
- Neue, H.U. 1993. Methane emission from rice fields. Bioscience
43:466–475.[Web of Science]
- Sass, R.L., F.M. Fisher, Y.B. Wang, F.T. Turner, and M.F. Jund. 1992. Methane emissions from rice fields: The effect of flood water management. Global Biogeochem. Cycles
6:249–262.
- Soil Survey Staff. 1975. Soil taxonomy: A basic system of soil classification for making and interpreting soil surveys. USDA-SCS Agric. Handb. 436. U.S. Gov. Print Office, Washington, DC.
- Trolldenier, G. 1995. Methanogenesis during rice growth as related to the water regime between crop season. Biol. Fertil. Soils
19:84–86.
- Wang, B., Y. Xu, Z. Wang, Z. Li, and Y. Ding. 1999. Methane production potentials of twenty-eight rice soils in China. Biol. Fertil. Soils.
29:74–80.
- Whalen, S.C., and W.S. Reeburgh. 1996. Moisture and temperature sensitivity of CH4 oxidation in boreal soils. Soil Biol. Biochem.
28:1271–1281.
- Xu, H., G.X. Xing, Z.C. Cai, and H. Tsuruta. 1997. Nitrous oxide emissions from three rice paddy fields in China. Nutrient. Cycling Agroecosyst.
49:23–28.
- Xu, H., Z.C. Cai, X.P. Li, and H. Tsuruta. 2000. Effect of antecedent soil water regime and rice straw application time on CH4 emission from rice cultivation. Aust. J. Soil Res.
38:1–12.
- Yagi, K., H. Tsuruta, K. Minami, P. Chairoj, and W. Cholitkul. 1994. Methane emissions from Japanese and Thai paddy fields. p. 41–53. In K. Minami et al. (ed.) CH4 and N2O: Global emissions and controls from rice fields and other agricultural and industrial sources. Yokendo Publishers, Tokyo.
- Yagi, K., K. Kumagai, H. Tsuruta, and K. Minami. 1995. Emission, production and oxidation of methane in a Japanese rice paddy field. p. 231–243. In R. Lal et al. (ed.) Soil management and greenhouse effect. Lewis Publishers, Boca Raton, FL.
- Yagi, K., H. Tsuruta, K. Kanda, and K. Minami. 1996. Effect of water management on methane emission from a Japanese rice paddy field: Automated methane monitoring. Global Biogeochem. Cycles
10:255–267.
- Yagi, K., H. Tsuruta, and K. Minami. 1997. Possible options for mitigating methane emission from rice cultivation. Nutrient. Cycling Agroecosyst.
49:213–220.
This article has been cited by other articles:
|
|
|
|
 
P. Smith, D. Martino, Z. Cai, D. Gwary, H. Janzen, P. Kumar, B. McCarl, S. Ogle, F. O'Mara, C. Rice, et al.
Greenhouse gas mitigation in agriculture
Phil Trans R Soc B,
February 27, 2008;
363(1492):
789 - 813.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
