Plant and Soil

, Volume 335, Issue 1, pp 373–383

Effects of soil water content and rice straw incorporation in the fallow season on CH4 emissions during fallow and the following rice-cropping seasons

Authors

  • Hua Xu
    • State Key Laboratory of Soil and Sustainable AgricultureInstitute of Soil Science, Chinese Academy of Sciences
    • Crop Production and Environment DivisionJapan International Research Center for Agricultural Sciences
Regular Article

DOI: 10.1007/s11104-010-0426-y

Cite this article as:
Xu, H. & Hosen, Y. Plant Soil (2010) 335: 373. doi:10.1007/s11104-010-0426-y

Abstract

Methane (CH4) emissions from paddy fields are believed to contribute to the greenhouse effect. Yet, in the literature, only a few reports are available on the effects of soil moisture regime and straw application in the non-rice-growing season separately on CH4 emissions during the rice-growing season. The objective of this study was to investigate CH4 emissions during the winter fallow and the following rice-growing season as affected by soil moisture regime and rice straw application during the fallow season. The experiment was designed to have 10 treatments, that is, five soil water contents (18%, 38%, 59%, and 79% of soil water-holding capacity [SWHC] and flooding; hereafter, W18, W38, W59, W79, and W100) and two rice straw application rates (0.91 and 4.55 g kg-1 dry soil; hereafter, Sl and Sh) during the fallow season. Both W100 and W79 showed obvious CH4 emissions during the fallow season, contributing 5.3% and 5.9% (Sl) and 34.8% and 27.8% (Sh), respectively, to their gross CH4 emissions, which increased significantly with the rising soil water content in the fallow season, except for W18. Rice straw application significantly affected gross CH4 emissions, but its effect was strongly influenced by soil moisture. The CH4 emissions per unit weight of rice straw applied of W38 and W59 were 9% and 16%, respectively, as much as that of W100. The findings demonstrate that keeping the soil water content in the range of 38–59% SWHC in the fallow season is important for a reduction in CH4 emissions.

Keywords

Methane (CH4) emissionsFallow season managementRice strawSoil water contentSoil properties

Introduction

Methane (CH4) is a key greenhouse gas with global warming potential (GWP) 25 times greater than that of carbon dioxide (CO2) on a 100-year scale (IPCC 2007a). The global atmospheric concentration of CH4 increased from about 0.715 ppm in the pre-industrial years to 1.774 ppm in 2005 (IPCC 2007a). Flooded rice soil is a major contributor to this increase because of its characteristic anaerobic condition. Global annual CH4 emissions from paddy fields are estimated to be in the range of 25 to 100 Tg (IPCC 2001) or 31 to 112 Tg (IPCC 2007b), etc. To have an accurate scientific assessment of the source strength of paddy fields in terms of CH4 emissions, further information is needed about the factors affecting emissions.

Water and organic matter management is crucial in regulating CH4 emissions from rice fields. In general terms, rice-based agriculture has two distinct seasons—the rice-growing season and non-rice-growing season. The effects of water and organic matter management during the rice-growing season on CH4 emissions have been intensively investigated (Cai 1997; Ma et al. 2008; Sass et al. 1992; Towprayoon et al. 2005; Watanabe et al. 1995; Yagi et al. 1996; Yan et al. 2005; Yu et al. 2004; Zou et al. 2005). Compared with continuous flooding, mid-season drainage and intermittent irrigation in the rice-growing season are effective ways to reduce CH4 emissions (Cai 1997; Sass et al. 1992; Towprayoon et al. 2005; Yagi et al. 1996; Yan et al. 2005; Zou et al. 2005). Incorporation of organic materials prior to rice cultivation was found to enhance CH4 emissions markedly (Cai 1997; Ma et al. 2008; Watanabe et al. 1995; Yan et al. 2005; Yu et al. 2004; Zou et al. 2005). In the non-rice-growing season, the water content of paddy soils can vary considerably, depending on climatic conditions and agricultural management, and rice straw is usually returned totally or partially (e.g., stubble) to a field after harvest. Soil water content and crop straw incorporation in the non-rice-growing season have also received some attention with respect to their roles in controlling CH4 emissions (Cai et al. 2000, 2003; Kang et al. 2002; Lu et al. 2000; Shiratori et al. 2007; Watanabe and Kimura 1998; Xu et al. 2000, 2003; Yagi et al. 1998; Yan et al. 2005). Soils flooded during the non-rice-growing season not only release substantial CH4 in that season (Cai et al. 2000, 2003; Yagi et al. 1998), but also have much higher CH4 emissions in the following rice-growing season than soils that are drained (Cai et al. 2003; Kang et al. 2002; Shiratori et al. 2007; Xu et al. 2000, 2003). The incorporation of crop straw earlier in the non-rice-growing season resulted in lower CH4 emissions in the following rice-growing season than incorporation right before rice cultivation (Lu et al. 2000; Watanabe and Kimura 1998; Xu et al. 2000). However, these studies mainly focused on flooded soils or the rice-growing season, and little on annual CH4 emissions as affected by soil water content and rice straw application rate during the non-rice-growing season.

To investigate the effects of soil water content and rice straw application during the non-rice-growing (fallow for this experiment) season on annual CH4 emissions, a pot experiment was carried out from December 2003 to September 2004.

Materials and methods

Experimental design

A pot experiment was carried out in a greenhouse at the experimental station of the Japan International Research Center for Agricultural Sciences (JIRCAS), Tsukuba, Ibaraki, Japan (N36.055°, E140.080°), under the conditions of a single-cropping rice farming system typical of the region. The soil used in the pots was gray lowland soil collected from the top 0.2-m layer of an experimental paddy field at the station, with properties listed as follows: pH 5.57, total C 15.18 g kg-1, and total N 1.38 g kg-1. A total of 30 pots (0.25 m in inner diameter and 0.35 m in height, each) were filled up each with 11 kg of the soil (oven dry-mass basis) on December 2, 2003. All the soils were left in fallow until May 19, 2004, and subjected separately to five moisture treatments: 18% (W18), 38% (W38), 59% (W59), and 79% (W79) of the soil water-holding capacity, and flooding (W100), during that season and two rice straw incorporation rates: 0.91 (Sl) and 4.55 (Sh) g kg-1 (returning rice straw partially and totally to the field, respectively). The intended soil water contents in the pots during the fallow season were maintained by making up for any loss with tap water every day based on the change in total weight of each pot. The experiment was fully randomized with three replicates.

On May 19, 2004, all the pots were flooded after having been incorporated with basal fertilizers consisting of 50 kg N ha-1 of urea, 80 kg P2O5 ha-1 of CaH2PO4, and 80 kg K2O ha-1 KCl. On May 21, 2004, one bunch of three seedlings of Koshihikari, the most popular variety of rice (Japonica) cultivated in Japan, was transplanted in each pot, at their 3- to 4-leaf stage. On July 31, 2004, the plants were top-dressed with 20 kg N ha-1 of urea and 20 kg K2O ha-1 of KCl. On September 11, 2004, the crop was harvested. The soils in all the pots were kept flooded with a water layer of at least 0.01-m thick throughout the rice-growing season, except for 58 h mid-season drainage and 4 d drainage right before the harvest.

Sampling and measuring

The closed chamber method was used to determine CH4 fluxes. Gas samples were collected at 6–15-d intervals in the fallow season and at 3–7-d intervals during the rice-growing season. A plexiglass chamber (0.26 m in inner diameter and 1 m in height) was placed over each pot with the bottom edge fitted right into the water trough (0.02 m in width and 0.05 m in depth) around the top edge of the pot, and gas samples were extracted through a three-port plastic valve at 10-min intervals (0, 10, 20, and 30 min) with a 10-ml plastic syringe. Gas concentrations were analyzed with a gas chromatograph (GC-14B, Shimadzu Corporation, Japan) equipped with a flame ionization detector for CH4 analyses. Methane fluxes were calculated by simple linear regression of chamber headspace concentration vs. sampling time.

During the rice-growing season, soil redox potential (Eh) was measured with Pt-tipped electrodes (Hirose Rika Co. Ltd., Japan) and an oxidation-reduction potential meter (PRN-41, Toa Corporation, Japan). For measuring soil Eh, electrodes were inserted into the soil at a depth of 0.1 m and kept there throughout the rice-growing season. Soil Eh was measured in triplicate. Soil temperatures at 0.1-m depth were measured with a hand-carried digital thermometer (Model 2455 03, Yokogawa Electric Corporation, Japan) during both the fallow and rice-growing seasons.

Soils pretreated in the fallow season were collected before rice transplantation for determination of soil C, NO3-, SO42-, and potentially reducible Fe and Mn. Soil C (including undecomposed straw debris) contents were analyzed with an NC analyzer (NC-900, Sumika Chemical Analysis Service, Japan). To determine concentrations of NO3- and SO42-, the soil samples were extracted by water and analyzed with an ion chromatograph (DX-120, Dionex Corporation, USA). Potentially reducible Fe and Mn were estimated from the difference between Fe and Mn extracted from the original soils and those incubated in 0.55 M dextrose for 30 d under an N2 atmosphere. The extractant used was ammonium acetate-acetic anhydride buffer (pH 2.8). Fe extract was determined with a colorimetric method using 1,10-phenanthroline, and Mn extract with a polarized Zeeman atomic absorption spectrophotometer (Z-8000, Hitachi Ltd., Japan).

Calculation of soil oxidation capacity

Soil oxidation capacity (OXC) was calculated using the equation reported by Zhang et al. (2009):
$$ OXC = 5\left[ {NO_3^{-} } \right] + 2\left[ {Mn\left( {IV} \right)} \right] + \left[ {Fe\left( {III} \right)} \right] + 8\left[ {SO_4^{2 - }} \right] $$
where brackets denote millimolar concentrations (mmol kg-1).

Statistical analysis

Statistical analysis was performed using SPSS 10.0 software for Windows (SPSS Inc., USA). Least significant difference tests were performed to compare means between treatments.

Results

Soil properties

The contents of main soil oxidants (NO3-, SO42-, and potentially reducible Fe and Mn), their integrated oxidation capacity (OXC), and soil reductant (C) in the soil collected before rice transplanting are listed in Table 1. Different from soil C content, which varied in a small range of 15.53–16.79, the contents of soil oxidants varied in a much greater range between different treatments. Though no obvious relationship was observed between soil water and C contents, treatments W18 and W79 were found to be higher than the other three soil water treatments in C content. Treatment Sh was generally higher in soil C content than treatment Sl (Table 1). The contents of soil oxidants decreased generally with the rising soil moisture content, with a few exceptions (NO3- contents in treatments W18 and W38, and SO42- content in treatment W79Sh). As the dominant soil oxidant, potentially reducible Fe contributed more than 51% to the OXC in the soil. The relationship between OXC and the antecedent soil water contents follows such a law: the higher the soil water contents in the fallow season, the lower the soil OXC, except for the treatment lowest in soil water content (Table 1). Each soil subjected to a moisture level during the fallow period with a higher rate of rice straw incorporation had almost the same OXC as its corresponding soil with a lower rate (Table 1).
Table 1

Contents of soil C (including undecomposed straw debris), NO3-, SO42-, potentially reducible Fe and Mn, and oxidation capacity in soils right before rice transplanting

Treatment

C (g kg-1)

NO3-(μg g-1)

SO42-(μg g-1)

Fe (μg g-1)

Mn (μg g-1)

OXC (mmol kg-1)

W18Sl

15.97 ± 0.22b

205 ± 10b

120 ± 12c

4068 ± 301d

218 ± 31d

107 ± 8e

W38Sl

15.60 ± 0.19a

496 ± 49c

97 ± 13c

4049 ± 160d

206 ± 25d

128 ± 9f

W59Sl

15.53 ± 0.14a

4 ± 2a

141 ± 18cd

3739 ± 203d

158 ± 22c

85 ± 6d

W79Sl

16.04 ± 0.26bc

2 ± 2a

166 ± 16d

2,278 ± 98c

52 ± 18b

57 ± 4c

W100Sl

15.66 ± 0.15a

0a

21 ± 3b

916 ± 40b

7 ± 4a

18 ± 1b

W18Sh

16.79 ± 0.39e

189 ± 63b

132 ± 20cd

4039 ± 286d

216 ± 20d

106 ± 13e

W38Sh

16.28 ± 0.24cd

651 ± 102d

120 ± 26c

4033 ± 250d

207 ± 19d

142 ± 16f

W59Sh

16.07 ± 0.33bc

3 ± 1a

150 ± 28cd

3688 ± 188d

156 ± 16c

84 ± 6d

W79Sh

16.47 ± 0.31de

1 ± 1a

313 ± 50e

1,812 ± 102c

31 ± 10b

60 ± 6c

W100Sh

15.69 ± 0.36a

0a

8 ± 2a

109 ± 18a

2 ± 2a

3 ± 1a

Values are means±standard deviation of three replicates; within a column, those followed by the same letter are not significantly different (P > 0.05)

W18, W38, W59, W79, and W100 stand for soil water content 18%, 38%, 59%, 79%, and 100% (flooded) of SWHC (soil water-holding capacity), respectively, in the fallow season; Sl and Sh for rice straw application rate 0.91 and 4.55 g kg-1, respectively, in the fallow season; Fe and Mn for potentially reducible Fe and Mn; and OXC for oxidation capacity

Temporal variations of CH4 flux, soil Eh, and soil temperature

Pronounced temporal variations of CH4 fluxes were observed for all 10 treatments (Fig. 1). More than 95% of the total CH4 was emitted between April and August of 2004 (Fig. 1). Significant CH4 emissions were observed in both treatments W100 and W79 in the late fallow season, but none in treatments W18, W38, and W59 during the whole fallow season (Fig. 1). Figure 1 also showed that CH4 emissions started at different times in different treatments. In treatments W100 and W79, CH4 emissions even occurred in the fallow season, while in treatments W18, W38, and W59, CH4 emissions began about 9–28 days after rice transplanting.
https://static-content.springer.com/image/art%3A10.1007%2Fs11104-010-0426-y/MediaObjects/11104_2010_426_Fig1_HTML.gif
Fig. 1

Temporal variations of CH4 flux during the fallow and following rice-growing seasons. W18, W38, W59, W79, and W100 stand for soil water content 18%, 38%, 59%, 79%, and 100% (flooded) of SWHC (soil water-holding capacity), respectively, in the fallow season; Sl and Sh for rice straw application rate 0.91 and 4.55 g kg-1, respectively, in the fallow season. Vertical bars indicate SDs of triplicates

The differences in soil Eh between the five water content treatments were quite obvious at the early stage of the rice-growing season (Fig. 2). Soil Eh as low as –110 mV or so was quickly established when rice was transplanted in treatments W100Sh and W100Sl, while in other treatments soil Eh values were much higher immediately after flooding and then decreased at various rates (Fig. 2). With the increase in water content during the fallow season, the post-flooding soil redox potential was kept lower in all cases, except for treatment W18, the lowest in soil moisture content (Fig. 2). Soil Eh of treatment W18 tended to decrease faster and became lower than those of treatments W38 and W59 within 10 days after rice transplanting (Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs11104-010-0426-y/MediaObjects/11104_2010_426_Fig2_HTML.gif
Fig. 2

Temporal variations of soil Eh during the rice-growing season. W18, W38, W59, W79, and W100 stand for soil water content 18%, 38%, 59%, 79%, and 100% (flooded) of SWHC (soil water-holding capacity), respectively, in the fallow season; Sl and Sh for rice straw application rate 0.91 and 4.55 g kg-1, respectively, in the fallow season. Vertical bars indicate SDs of triplicates

During the fallow season, soil temperature at a depth of 0.1 m of treatments W79 and W100, from which significant CH4 emissions were observed during this season, had a similar temporal variation pattern and fluctuated between 12.2 and 33.2°C. The temperature of 0.1-m-deep soil of all 10 treatments varied from 19.8 to 41.0°C during the rice-growing season (Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs11104-010-0426-y/MediaObjects/11104_2010_426_Fig3_HTML.gif
Fig. 3

Temporal variations of soil temperature at a depth of 0.1 m during the fallow and following rice-growing seasons. W79 and W100 stand for soil water content 79% and 100% (flooded) of SWHC (soil water-holding capacity), respectively, in the fallow season. FS and RS for the fallow season and rice-growing season, respectively

Seasonal total CH4 emissions and mean soil Eh

Treatment W100 was the highest in total CH4 emissions, followed in decreasing order by treatments W79, W18, W59, and W38. In contrast, treatment W100 was the lowest in mean soil Eh during the rice-growing season, followed in increasing order by treatments W79, W18, W59, and W38 (Table 2). These results indicated that CH4 emissions increased but soil Eh decreased significantly as soil water content increased during the fallow season, except in the case of the driest water condition, W18.
Table 2

Total CH4 emissions, CH4 emissions per unit weight of rice straw applied, and mean soil Eh in the rice-growing season

Treatment

Total CH4 emissions (mg CH4)

Percentage of NR to T

CH4 emissions per unit weight of rice straw applied (g CH4 g-1 dry matter)

Mean soil Eh in the rice-growing season (mV)

Non-rice- growing season (NR)

Rice-growing season (R)

Total (T)

W18Sl

0a

805 ± 148c

805 ± 148c

0

0.080

-21 ± 2d

W38Sl

0a

304 ± 32a

304 ± 32a

0

0.030

63 ± 39e

W59Sl

0a

604 ± 188bc

604 ± 188bc

0

0.060

6 ± 52de

W79Sl

106 ± 63b

1,699 ± 205d

1,805 ± 267d

5.9

0.180

-71±43c

W100Sl

147 ± 25b

2,625 ± 241e

2,772 ± 247e

5.3

0.277

-147±20a

W18Sh

0a

2,349 ± 311e

2,349 ± 311de

0

0.047

-83 ± 31bc

W38Sh

0a

486 ± 126b

486 ± 126b

0

0.010

30 ± 45de

W59Sh

0a

728 ± 113c

728 ± 113c

0

0.015

-1 ± 41de

W79Sh

890 ± 140c

2,309 ± 276e

3,199 ± 292e

27.8

0.064

-103 ± 22b

W100Sh

2,515 ± 432d

4,718 ± 462f

7,233 ± 133f

34.8

0.145

-157 ± 14a

Values are means±standard deviation of three replicates; within a column, those followed by the same letter are not significantly different (P > 0.05)

W18, W38, W59, W79, and W100 stand for soil water content 18%, 38%, 59%, 79%, and 100% (flooded) of SWHC (soil water-holding capacity), respectively, in the fallow season; Sl and Sh for rice straw application rates 0.91 and 4.55 g kg-1, respectively, in the fallow season

Soils of treatments W79 and W100 not only emitted a large amount of CH4 during the rice-growing season, but also released a substantial amount of CH4 in the fallow season. CH4 emissions during the fallow season contributed 5.3–34.8% to the total CH4 emissions (Table 2). As Table 2 shows, the percentage values of treatments W79Sh and W100Sh were much larger than those of treatments W79Sl and W100Sl.

Rice straw incorporated at a high rate increased CH4 emissions significantly by a factor of 2.92, 2.61, 1.77, and 1.60 in treatments W18, W100, W79, and W38, respectively, and insignificantly by a factor of 1.21 in treatment W59 (Table 2). The amount of CH4 emissions per unit weight of rice straw incorporation was the lowest in treatment W38, followed by treatment W59, being only 9% and 16% of that in treatment W100 (Table 2). Soils incorporated with a high rate of rice straw were relatively lower in mean soil Eh (Table 2).

Relationship of CH4 emissions with soil Eh and temperature

As is shown in Figs. 1 and 2, the difference between treatments in temporal variation pattern of soil Eh can well explain the difference between treatments in temporal variation pattern of CH4 flux. The quicker the decrease in soil Eh, the earlier CH4 emissions began after rice transplanting. A significant negative linear correlation was observed between CH4 flux and soil Eh during the rice-growing season in treatments W38 and W59 (Table 3), but it was non-significant in treatments W18, W79, and W100, indicating that soil Eh is not a critical factor controlling the seasonal variation of CH4 fluxes in the three treatments.
Table 3

Correlation coefficients of CH4 fluxes with soil Eh and soil temperature in linear regression

Treatment

CH4 flux—soil Eh

CH4 flux—soil temperature

Rice season

Fallow season

Rice season

W18Sl

-0.243

0.285

W38Sl

-0.620**

0.257

W59Sl

-0.545*

0.364

W79Sl

-0.076

0.581*

0.517*

W100Sl

-0.034

0.499*

0.436*

W18Sh

-0.126

0.272

W38Sh

-0.478*

0.294

W59Sh

-0.595**

0.323

W79Sh

-0.283

0.692**

0.552**

W100Sh

-0.057

0.747**

0.540*

**Significant at P < 0.01; *Significant at P < 0.05

W18, W38, W59, W79, and W100 stand for soil water content 18%, 38%, 59%, 79%, and 100% (flooded) of SWHC (soil water-holding capacity), respectively, in the fallow season; Sl and Sh for rice straw application rate 0.91 and 4.55 g kg-1, respectively, in the fallow season

The relationship between soil temperature and CH4 flux seemed to be generally contrary to the case of soil Eh. CH4 flux had a significant positive linear relationship with soil temperature during the rice-growing season in treatments W79 and W100, while it was not significant in treatments W18, W38, and W59 (Table 3). Significant correlations were also found between soil temperature and CH4 flux during the fallow season in treatments W79 and W100 (Table 3).

Discussion

Methane production is the terminal step in anaerobic microbial decomposition of organic substances and hence requires a sufficiently reduced soil condition (low Eh). The activity of methanogenic bacteria, the decomposition rate of soil organic matter, and production and transport rate of CH4 all increase with soil temperature. Methane emissions from rice fields are simultaneously influenced by a number of agro-environmental factors, among which soil Eh and temperature are the two most commonly measured factors. In treatments W38 and W59, soil Eh began to decrease after rice transplanting slowly from a high value (288–500 mV) down to below 0 mV (Fig. 2) in 40 days. On the other hand, soil temperature fluctuated within a suitable range for CH4 production during the rice-growing season (Fig. 3). Therefore, soil Eh was more likely than soil temperature to be the factor controlling CH4 emissions and their seasonal variation in treatments W38 and W59. Hence, no significant correlation between CH4 fluxes and soil temperatures was observed, but the relationship between CH4 fluxes and soil Eh was found to be significant in these two treatments during the rice-growing season (Table 3). In treatments W79 and W100, soil Eh remained below 0 mV during most of the rice-growing season, and hence it may not have been a limiting factor for CH4 production (Fig. 2) and seasonal variation of CH4 fluxes may have been mainly controlled by other factors such as soil temperature. Therefore, a significant relationship was observed between CH4 fluxes and soil temperature, not between CH4 fluxes and soil Eh, during the rice-growing period in treatments W79 and W100 (Table 3). A significant relationship between CH4 flux and soil temperature was also found during the fallow season in treatments W79 and W100 (Table 3), which is in agreement with the findings of the field experiment in Chongqing, China (Cai et al. 2003).

Since methanogenic bacteria can metabolize organic substances only in strict anaerobic conditions, it is reasonable to deem soil Eh as the most important factor influencing CH4 production in paddy fields. An insignificant relationship between CH4 fluxes and soil Eh was observed during the rice-growing season in treatments W79 and W100, which did not contradict but instead further proved the important role soil Eh plays in affecting CH4 production (Table 3). What soil Eh failed to influence significantly was only the temporal variation of CH4 fluxes in treatments W79 and W100, but not CH4 flux itself. As Table 2 shows, the total CH4 emissions were significantly related to the mean soil Eh during the rice-growing season (r = -0.92, p < 0.01). The process of soil reduction is controlled by the relative abundance of electron donors and electron acceptors in the soil. In the absence of O2, the main electron-accepting chemical species are NO3-, potentially reducible Fe and Mn, and SO42-, and the main electron donor is readily decomposable organic matter (Yagi et al. 1995). The formation of CH4 occurs only after NO3-, Mn4+, Fe3+, and SO42- are reduced in thermodynamical sequential order by soil organic matter (Neue 1993).

Soil C not only helps stimulate soil reduction and create a strict reductive condition for CH4 production but also acts as a substrate for CH4 production. Therefore, soil C should negatively affect soil Eh and positively affect CH4 emissions. However, the correlation coefficient between soil Eh and soil C was only -0.12 (p = 0.89) and between CH4 emissions and soil C was 0.048 (p = 0.89) during the rice-growing season, showing no significant effect of soil C. This is probably because the effect of soil C was overshadowed by the strong effect of soil OXC on soil Eh (r = 0.87; p = 0.001) and CH4 emissions (r = -0.82; p = 0.003). Zhang et al. (2009) also found that soil OXC was more important than soil C in determining redox potential of forest soil. The insignificant correlation between CH4 emissions and soil C was once reported by Huang et al. (2002) in a pot experiment, though a significant correlation between CH4 production and soil C was found in some incubation studies (Crozier et al. 1995; Wang et al. 1999). The contribution of soil C versus C from rice plants to CH4 emissions might be very small in the presence of a rice crop (Huang et al. 2002).

Studies on the effect of soil water regime during the non-rice-growing season on annual CH4 emissions have focused mainly on flooded soils (Cai et al. 2000, 2003; Yagi et al. 1998). Methane emissions during the fallow season as affected by soil moisture regime were rarely reported. Though unsaturated and with soil OXC still as high as 57 (Sl) and 60 (Sh) mmol kg-1 before rice transplantation, soils with 79% of SWHC (soil water-holding capacity) were found unexpectedly releasing a substantial amount of CH4 during the fallow season, in particular when they were incorporated with a high rate of rice straw (Tables 1 and 2). These results indicate that there must have been a considerable amount of OXC-depleted anaerobic microsites in soils of treatment W79 during the fallow season. The fact that the contribution of CH4 emissions during the fallow season to the total of the two observation seasons in treatment W79 was similar to that in treatment W100 further demonstrated the importance of soils with 79% of SWHC in emitting CH4 to the atmosphere (Table 2).

In this experiment, only CH4 emissions during the fallow period from Dec. 2 to May 19 were measured. Under actual field conditions in Ibaraki, Japan, the fallow season was longer and lasted for nearly eight months from September to May. Moreover, in contrast to the finding that CH4 emissions started from early March, CH4 emissions from soils with 79% or higher SWHC may well occur at the very beginning of a fallow season, during which the temperature is higher than in winter around Dec. to Feb., under field conditions; hence, CH4 emissions from paddy fields during an actual fallow season might be higher than what was obtained in this experiment, as was confirmed by a lysimeter experiment (Yagi et al. 1998) and a field experiment (Cai et al. 2003). Yagi et al. (1998) found that CH4 emissions during the fallow periods (from September to May, with rice stubbles standing 5 cm above ground) were equivalent to 14–18% of those during the previous rice cultivation period if the paddies were continuously flooded year-round. The contribution of CH4 emissions during the winter crop season (from September to May, with rice stubbles standing 30 cm above ground) to annual CH4 emissions was as high as 39.4–44.3% (Cai et al. 2003). These results suggest that CH4 emissions from paddy soil during the fallow or non-rice-growing seasons of any other kind deserve more attention for a more reliable estimation of CH4 emissions.

Methane emissions in the rice-growing season increased significantly with rising soil water content in the fallow season, except for the treatment lowest in soil water content, treatment W18 (Table 1). This finding was consistent with a previous report (Xu et al. 2003). Soil water regime in the preceding season significantly affected the contents of soil C and the integrated value of soil oxidants (OXC) before rice transplanting (Table 1) and consequently soil Eh and CH4 emissions in the rice-growing season (Table 2). Treatment W18 was significantly lower in soil OXC content than treatment W38, but significantly higher in soil C content than treatments W38 and W59, and consequently significantly higher in CH4 emissions than treatments W38 and W59 (Tables 1 and 2). The fact that OXC was lower and soil C content was higher in soils with 18% of SWHC than in soils with 38% of SWHC (Table 1) could be explained by water stress on activities of soil microbes. Re-oxidation of paddy soil reduced during the rice-growing season would be stimulated by the activities of soil microbes if they were not under stress. Since the water content, 18% of SWHC, was very likely lower than the wilting point of the tested soil, it could be assumed that soil microbes were under stress and therefore the decomposition of soil organic matter and oxidation of soil reductants were inhibited at least partially in soils with 18% of SWHC compared with those in soils with more than 18% of SWHC. Higher soil water contents (W79 and W100) caused higher CH4 emissions during both the fallow and the following rice-growing seasons, whereas lower soil water content (W18) led to higher CH4 emissions during the subsequent rice-growing season (Table 2), which indicates that CH4 emissions from paddy fields would be reduced to a minimum if soil water content could be adjusted to a suitable level, around 38–59% of SWHC during the fallow season.

Incorporation of crop straw in the non-rice-cropping season instead of the rice-growing season is recommended as an option to reduce CH4 emissions from rice fields (Lu et al. 2000; Watanabe and Kimura 1998; Xu et al. 2000). However, studies on the effects of prior application of crop straw on CH4 emissions have focused mainly on the effects during the rice-growing season, and didn’t consider the effects of soil moisture (Lu et al. 2000; Watanabe and Kimura 1998; Xu et al. 2000). Methane emissions during the rice-growing season increased with the increase in rice straw application rate in the fallow season (Table 2), which was consistent with previous reports (Lu et al. 2000; Watanabe and Kimura 1998). However, the stimulating effect of rice straw application on CH4 emissions was greatly influenced by soil water content in the fallow season. The stimulating factor of rice straw application on CH4 emissions was the smallest, being only 1.21 in treatment W59, followed by 1.60 (W38), 1.77 (W79), 2.61 (W100), and 2.92 (W18). On the other hand, treatments W38 and W59 were the lowest and the second lowest in CH4 emissions per unit weight of rice straw applied, while treatments W100 and W79 were the highest and the second highest (Table 2). Moreover, the contribution of CH4 emissions during the fallow season to the total from the two seasons, the fallow and the following rice-growing season, increased from 5.3–5.9% to 27.8–34.8% when the rice straw incorporation rate increased from 0.91 to 4.55 g kg-1 in treatments W79 and W100 (Table 2). All these findings indicate that rice straw incorporation should be avoided when soil water content is above 79% of SWHC, in particular when soil is flooded, and recommended when soil water content is around 38–59% of SWHC in a fallow season to effectively reduce the stimulating effect of rice straw application on CH4 emissions.

Acknowledgments

Support for this research was provided by the Japan International Research Center for Agricultural Sciences (JIRCAS) through a fellowship award to H. Xu. We acknowledge the support of the Crop Production and Environment Division of JIRCAS throughout the conduct of this work.

Copyright information

© Springer Science+Business Media B.V. 2010