Interactive comment on “ Pathway of CH 4 production , fraction of CH 4 oxidized , and 13 C isotope fractionation in a straw incorporated rice field ”

Straw incorporation generally increases CH 4 emission from rice fields, but its effects on the mechanism of CH 4 emission, especially on the pathway of CH 4 production and the fraction of CH 4 oxidized, are not well known. To investigate the methanogenic pathway, the fraction of CH 4 oxidized as well as the stable carbon isotope fractionation during the oxidation and transport of CH 4 as affected by straw incorporation, observations were conducted of production and oxidation of CH 4 in paddy soil and rice roots and δ 13 C-values of produced CH 4 and CO 2 , and emitted CH 4 in incubation and field experiments. Straw incorporation significantly enhanced CH 4 production potentials of the paddy soil and rice roots. However, it increased the relative contribution of acetate to total CH 4 production ( F ac ) in the paddy soil by ∼10–30%, but decreased F ac -value of the rice roots by ∼5–20%. Compared with rice roots, paddy soil was more important in acetoclastic methanogenesis, with F ac -value being 6–30% higher. Straw incorporation highly decreased the fraction of CH 4 oxidized ( F ox ) by 41–71%, probably attributed to the fact that it increased CH 4 oxidation potential whereas CH 4 production potential was increased to a larger extent. There was little CH 4 formed during aerobic incubation, and the produced CH 4 was more 13 C-enriched relative to that of anaerobic incubation. Assuming δ 13 C-values of CH 4 aerobically produced in paddy soil to be the δ 13 C-values of residual CH 4 after being oxidized, ( F ox -value still appeared to be 45–68% lower when straw was incorporated. Oxidation fractionation factor (α ox ) was higher with straw incorporation (1.033) than without straw incorporation (1.025). The δ 13 C-values of CH 4 emitted after cutting of the plants (−50 to −43‰) were more positive than those of before (−58 to −55‰), suggesting a transport fractionation factor (v transport ) was −8.0‰ with straw incorporation and −12.0‰ without straw incorporation. Causes of this difference may be related to the diffusion process in transport as affected by growth of rice plants and pressure in the rhizosphere. The experiment shows that straw incorporation increases the contribution of acetate to total methanogenesis in paddy soil but decreases it on rice roots, and it significantly decreases the fraction of CH 4 oxidized in the field and expands oxidation fractionation while reducing transport fractionation.

Abstract.Straw incorporation generally increases CH 4 emission from rice fields, but its effects on the mechanism of CH 4 emission, especially on the pathway of CH 4 production and the fraction of CH 4 oxidized, are not well known.To investigate the methanogenic pathway, the fraction of CH 4 oxidized as well as the stable carbon isotope fractionation during the oxidation and transport of CH 4 as affected by straw incorporation, observations were conducted of production and oxidation of CH 4 in paddy soil and rice roots and δ 13 C-values of produced CH 4 and CO 2 , and emitted CH 4 in incubation and field experiments.Straw incorporation significantly enhanced CH 4 production potentials of the paddy soil and rice roots.However, it increased the relative contribution of acetate to total CH 4 production (F ac ) in the paddy soil by ∼ 10-30 %, but decreased F ac -value of the rice roots by ∼ 5-20 %.Compared with rice roots, paddy soil was more important in acetoclastic methanogenesis, with F ac -value being 6-30 % higher.Straw incorporation highly decreased the fraction of CH 4 oxidized (F ox ) by 41-71 %, probably attributed to the fact that it increased CH 4 oxidation potential whereas CH 4 production potential was increased to a larger extent.There was little CH 4 formed during aerobic incubation, and the produced CH 4 was more 13 C-enriched relative to that of anaerobic incubation.Assuming δ 13 C-values of CH 4 aerobically produced in paddy soil to be the δ 13 Cvalues of residual CH 4 after being oxidized, F ox -value still appeared to be 45-68 % lower when straw was incorporated.Oxidation fractionation factor (α ox ) was higher with straw incorporation (1.033) than without straw incorporation (1.025).The δ 13 C-values of CH 4 emitted after cutting of the plants (−50 to −43 ‰) were more positive than those of before (−58 to −55 ‰), suggesting a transport fractionation factor (ε transport ) was −8.0 ‰ with straw incorporation and −12.0 ‰ without straw incorporation.Causes of this difference may be related to the diffusion process in transport as affected by growth of rice plants and pressure in the rhizosphere.The experiment shows that straw incorporation increases the contribution of acetate to total methanogenesis in paddy soil but decreases it on rice roots, and it significantly decreases the fraction of CH 4 oxidized in the field and expands oxidation fractionation while reducing transport fractionation.

Introduction
Atmospheric methane (CH 4 ), the second most important greenhouse gas next to carbon dioxide (CO 2 ), reached 1808 nL L −1 in 2010 (WMO, 2010).As an important source of anthropogenic CH 4 , paddy field is responsible for ∼ 5-19 % of the global CH 4 emission (IPCC, 2007).China, being one of the most important rice-producing countries in the world, has an abundant straw resource, which reaches 6.4 × 10 8 t yr −1 and reached even up to 7.3 × 10 8 t in 2010 (Xiong et al., 2010).Straw incorporation is regarded as a key practice in management of organic fertilizer in rice cultivation.And it proves to be able to gradually improve soil structure (Wang et al., 2010), increase soil organic carbon (Deng et al., 2010;Ma et al., 2010;Wang et al., 2010), raise soil fertility (Deng et al., 2010;Ma et al., 2010;Wang et al., 2010), and in the long run promote the agroecosystem into a benign cycle.However, in the short term, it highly increases CH 4 emission from the paddy fields (Cai, 1997;Jiang et al., 2003;Ma et al., 2009), causes microbial immobilization of soil mineral N (Tanaka et al., 1990;Jensen, 1997), and accumulates organic acids (Tanaka et al., 1990;Shan et al., 2006), thus affecting growth of the crop.Incorporation of straw significantly increases CH 4 emission from paddy fields, which has been considerably reported (Jiang et al., 2003;Ma et al., 2008Ma et al., , 2009)), but its effects on production and oxidation of CH 4 are not clear yet.
In paddy fields, CH 4 is an important end product of the degradation of organic matter under anaerobic conditions (Cicerone and Oremland, 1988;Conrad, 2007).Organic matter is fermented into acetate, CO 2 , H 2 , propionate as well as other fatty acids.However, acetate, CO 2 and H 2 are the main substrates that methanogenic bacteria use for production of CH 4 (Krüger et al., 2002;Conrad et al., 2010).The relative contribution of acetoclastic (CH 3 COOH → CH 4 + CO 2 ) and hydrogenotrophic (CO 2 +4H 2 →CH 4 +2H 2 O) methanogenesis to total CH 4 formation is sensitive to availability of the substrates and varies with the growth of rice plants (Whiticar et al., 1986;Conrad, 1999;Krüger et al., 2002).Degradation of straw generates abundant organic acids, such as formic, acetic, propionic and butyric acids (Shan et al., 2006), which is likely to increase the contribution of acetoclastic methanogenesis.Moreover, it provides methanogenic bacteria with plenty of precursors to produce CH 4 and accelerates the decline of soil Eh, thus forming a favorable environment condition for growth of methanogens, which in turn promotes formation of CH 4 and further affects CH 4 oxidation capacity in the field (Bender and Conrad, 1995;Arif et al., 1996).On the other hand, rice roots themselves not only excrete organic acid and slough off old or dead tissues as sources of carbon and energy for CH 4 production, but also act as an important site where CH 4 is oxidized by oxygen available from root secretion in the rhizosphere.Therefore, the growth of rice roots probably stimulates growth and activity of methanotrophs, and consequently increase the potential of CH 4 oxidation.A considerable number of studies have reported production and oxidation of CH 4 in paddy soil or on rice roots (Frenzel and Bosse, 1996;Bosse and Frenzel, 1997;Conrad and Klose, 1999;Lehmann-Richter et al., 1999;Dan et al., 2001;Krüger et al., 2002;Zhang et al., 2011c), but little has focused on the effect of straw incorporation on CH 4 production and CH 4 oxidation separately, let alone on pathway of CH 4 production and fraction of CH 4 oxidized.
In the study of CH 4 production and oxidation processes in the rice ecosystem, the stable carbon isotope technique, deemed to be a feasible and very effective method, has been widely used (Sugimoto and Wada, 1993;Chanton et al., 1997;Tyler et al., 1997;Bilek et al., 1999;Krüger et al., 2002;Conrad and Klose, 2005).Relative contribution of the two methanogenic pathways and proportion of CH 4 oxidized in the field can be estimated if δ 13 C-values of CH 4 , CO 2 and acetate are measured and relevant fractionation factors (ε acetate , α CO 2 /CH 4 , α ox and ε transport ) are known.The carbon isotope fractionation factors of CH 4 oxidation (α ox ) and CH 4 transport (ε transport ) are crucial to the quantification of the fraction of CH 4 oxidized in applying the stable carbon isotope technique (Tyler et al., 1997;Bilek et al., 1999;Krüger et al., 2002;Zhang et al., 2009).Therefore, the acquisition of reliable and exact fractionation factors (α ox and ε transport ) is an important precondition for use of the method in studying CH 4 oxidation.Usually a high proportion of the CH 4 produced in the paddy fields is oxidized in the rhizosphere and at the soil-water interface.The remaining CH 4 is emitted from the soil to the atmosphere mainly through the aerenchyma of rice plants.Due to the fact that 12 CH 4 is consumed faster than 13 CH 4 by methane-oxidizing bacteria as well as 12 CH 4 transports faster than 13 CH 4 in the process of CH 4 transport (Whiticar, 1999;Chanton, 2005;Venkiteswaran and Schiff, 2005), significant isotopic fractionation is observed during these processes.Previous studies mainly focused on fractionation factor α ox in the landfill cover soils (see Chanton et al., 2008a, b), rather than in paddy soil (Krüger et al., 2002).Moreover, early reports showed that fractionation factor ε transport varied with rice growth and with rice variety as well (Chanton et al., 1997;Tyler et al., 1997;Bilek et al., 1999;Krüger et al., 2002).Further study is necessary to discuss the effect of straw incorporation on growth of rice plants and eventually on fractionation factors, α ox of paddy soil and ε transport of rice plants.
In recent studies, CH 4 concentration in soil solution of the field, δ 13 C-values of CH 4 emitted before and after cutting of the plants, production and oxidation of CH 4 in paddy soil and fresh rice roots, and δ 13 C-values of CH 4 and CO 2 produced in anaerobic and aerobic incubations were measured in field and incubation experiments to clarify the effect of straw incorporation on CH 4 production and oxidation, respectively, especially on pathway of CH 4 production and fraction of CH 4 oxidized in the field.Moreover, stable carbon isotope fractionation factors α ox and ε transport were investigated to exactly estimate the effect of straw incorporation on them and finally on the fraction of CH 4 oxidized.

Design of the experiment
The experimental field was located at Baitu Town, Jurong City, Jiangsu Province, China (31 • 58 N, 119 • 18 E).The main characteristics of this site have been described in detail before (Zhang et al., 2012).Two treatments, WS (straw incorporation) and CK (without straw incorporation), were designed with experimental plots in triplicate.Stubble and weeds were all removed from the experimental plots.Then, dry wheat straw (C / N = 85, δ 13 C org = −28.3‰), chopped to ∼ 10 cm in length, was evenly spread over the plots of Treatment WS, and raked into the topsoil (0-15 cm).Rice seedlings (Huajing 3), 32-days-old, were transplanted on 26 June in all plots, and the crop was harvested on 3 November.The same fertilization and water management practices were adopted in the experiment as in the local rice cultivation.For details, please refer to Table 1.

Field experiment
Soil solution was collected from each plot using a Rhizon soil moisture sampler (Zhang et al., 2012).Prior to sampling, about 5 mL of soil solution were extracted using 18 mL vacuum vial to flush and purge the sampler.Approximate 10 mL of water were then drawn into another vial for further analysis.CH 4 concentration in the headspace of the vial was measured on a GC-FID.The CH 4 concentration (C CH4 ) in soil solution was calculated using where m stands for mixing ratio of CH 4 in the headspace of a vial (µL L −1 ), M V for gas volume of an ideal gas (24.78L mol −1 at 25 • C), G V for volume of the gas headspace of the vial (L), and G L for volume of liquid in the vial (L).Simultaneously, soil redox potential (Eh) at the depth of 10 cm was measured, using Pt-tipped electrodes (Hirose Rika Co. Ltd., Japan) and an oxidation-reduction potential meter with a reference electrode (Toa PRN-41).Soil temperature at the depth of 10 cm was measured with a hand-carried digital thermometer (Yokogawa Meters & Instruments Corporation, Japan).Triplicate soil cores were collected from each experimental plot using a stainless steel corer with an inner diameter of 7 cm and a length of 25 cm (Zhang et al., 2011a) and then prepared into mixture on a plot basis.Samples from the mixture, each about 50 g (dry weight), were promptly taken and transferred into the 250 mL Erlenmeyer flask separately, and turned into slurries with N 2 -flushed de-ionized sterile water in the ratio of 1 : 1 (soil/water).During the whole process, the samples were constantly flushed with N 2 to remove O 2 and CH 4 , and the flasks containing these samples were then sealed for anaerobic incubation.Other flasks with air headspace were sealed directly for aerobic incubation.All flasks were sealed with rubber stoppers fitted with silicon septum that allowed sampling of headspace gas.Finally, they were stored in N 2 at 4 • C and transported back to the lab as soon as possible for further analysis.A portion of soil sample was dried for 72 h at 60 • C for determination of isotopic composition of the organic carbon.
Rice plants, complete with roots, were carefully collected from the plots at each of the four rice growth stages: tillering stage (TS, 16 July), booting stage (BS, 15 August), grainfilling stage (FS, 22 September), and ripening stage (RS, 12 October) (Zhang et al., 2011c).The roots were washed with N 2 -flushed demineralized water and cut off from the green shoots at a point, 1-2 cm from the root with a razor blade.The fresh roots, 20 g each portion, were then put into flasks.Further preparation and processes of the roots were the same as for the soil and detailed in the preceding paragraph.The shoots were dried at 60 • C for 72 h for dry weight measurement, and then stored at room temperature for determination of isotopic composition of the organic carbon.Rice grain yield was measured at harvest.
Gas samples of emitted CH 4 were taken simultaneously before and after the plants were cut at the late booting stage (28 August) using specially designed PVC bottomless pots for determining stable carbon isotope fractionation during CH 4 transport (ε transport ) through the aerenchyma of rice plants (Zhang et al., 2013).The pot, 30 cm in height and 17 cm in diameter, was designed to have a water-filled trough around its top, allowing a chamber to rest on the pot with the joint completely sealed to avoid any possible gas exchange during the sampling times.Each plot had two such pots installed ∼ 7-10 days before gas sampling began, in a way to keep the plants in the center of each pot.A PVC plate (18 cm in diameter) with a hole (4 cm in diameter) in the center was placed on top of each pot, allowing the plant to grow through the hole and keep divided into two parts.Then, one plant inside the pot was cut right above the plate while the other remained intact as the control.Finally, chambers (30 × 30 × 100 cm) were laid on the pots, and gas samples in the headspace of the chambers were collected simultaneously using 500 mL bags (aluminium foil compound membrane, Delin Gas Packing Co., Ltd, Dalian, China) with a small battery-driven pump for measurement of δ 13 C-value of the emitted CH 4 (Zhang et al., 2011b).

Incubation experiment
Potential CH 4 production rates in paddy soil and rice roots were determined anaerobically (Zhang et al., 2011c).Flasks used for anaerobic incubation were flushed with N 2 consecutively six times through double-ended needles connecting a vacuum pump to purge the air in the flasks of residual CH 4 and O 2 .Simultaneously, CH 4 production was determined aerobically and flasks with air headspace were used directly in the experiment.They were then incubated at a temperature the same as measured in the field for 50 h in darkness.Gas samples were collected twice with a pressure lock syringe, 1 h and 50 h later after the flasks were heavily shaken by hand, and analyzed for CH 4 on a GC-FID.CH 4 production was calculated using the linear regression of CH 4 increasing with the incubation time.
Potential CH 4 oxidation rates in paddy soil and rice roots were determined aerobically, using the same equipments as described above but with air headspace in the flasks.Into each flask pure CH 4 was injected to make a high concentration inside (∼ 10 000 µL L −1 ).Then the flasks were incubated in dark under the same temperature as in the field and shaken at 120 rpm.CH 4 depletion was measured by sampling the headspace gas in the flask after vigorously shaking for subsequent GC-FID analysis.The first sample was collected generally 30 min after pure CH 4 was injected and made homogeneously distributed inside the flask.Samples were then taken in 2-3 h intervals during the first 8 h of the experiment.The flasks were left over night and sampled the next day in 2 h intervals again.CH 4 oxidation was calculated using the linear regression of the CH 4 depletion with incubation time.

Analytical methods
CH 4 concentration in gas samples was analyzed with a gas chromatograph (Shimadzu GC-12A, Kyoto, Japan) equipped with a flame ionization detector (FID).δ 13 C-values were determined with a Finnigan MAT-253 isotope ratio mass spectrometer (IRMS, Thermo Finnigan, Bremen, Germany) using the continuous flow technique.The IRMS had a fully automatic interface for pre-GC concentration (Pre-Con) of trace gases (Cao et al., 2008;Zhang et al., 2011b).Isotope ratios were expressed in the standard delta notation: of sample and standard, respectively.Preci-sion of the repeated analyses was ±0.196 ‰ (n = 9) with 2.02 µL L −1 CH 4 injected.Samples of dried soil and plants were analyzed for carbon isotope composition with a Finnigan MAT-251 isotope ratio mass spectrometer (Thermo Finnigan, Bremen, Germany).

Statistical analysis
Statistical analysis was done using the SPSS 18.0 software for Windows (SPSS Inc., Chicago).Least significant difference (LSD) tests were used to compare means between treatments.Standard deviation of the means was calculated using the Microsoft Excel 2003 software for Windows.

CH 4 concentration in soil solution
CH 4 concentration in soil solution of the two treatments dropped sharply over the season, from as high as 150-300 µmol L −1 at the tillering stage to ∼ 1 µmol L −1 at the ripening stage (Fig. 1a).Obviously, CH 4 concentration in soil solution was much higher in Treatment WS (180-330 µmol L −1 ) than in Treatment CK (40-140 µmol L −1 ) at the tillering and booting stages.However, no significant discrepancy between the two treatments was observed at the grain-filling and ripening stages (Fig. 1a).Similarly, soil Eh was significantly lower in Treatment WS (−200 to −150 mV) than in Treatment CK (−50 to −10 mV) at the tillering and booting stages while no obvious difference was detected between the two treatments at the grain-filling and ripening stages (Fig. 1b).Soil temperature ranged from 17.6 • C to 29.7 • C during the whole season, with an average of 25.1 • C.

CH 4 production potential in paddy soil and rice roots under anaerobic incubation
The CH 4 production potentials of paddy soil and rice roots in the two treatments (Fig. 2a, d) varied in similar patterns, rising up to the peak (soil: 0.6-1.4µg CH 4 g −1 d −1 ; roots: 18.5-49.4µg CH 4 g −1 d −1 ) at the booting stage, and declined rapidly towards the valley at the ripening stage.The potential was significantly higher in Treatment WS than in Treatment CK at the booting stage (P < 0.05).As a whole, straw incorporation increased CH 4 production potential in paddy soil and rice roots obviously, with the mean increased by 95 % and 134 % relative to those in Treatment CK, respectively (Fig. 2a, d).
With the growth of rice plants, δ 13 C-value of the CH 4 produced in paddy soil became more positive in both treatments (Fig. 2b).The produced CH 4 was more 13 C-enriched in Treatment WS (−70 to −47 ‰) than in Treatment CK (−75 to −56 ‰) at all four rice growth stages (Fig. 2b), with δ 13 C-values being 5-13 ‰ higher in Treatment WS than in Treatment CK.On the contrary, CH 4 produced on rice roots was gradually becoming 13 C-depleted from the tillering stage to the ripening stage, and it was relatively more obvious in Treatment WS (−83 to −70 ‰) than in Treatment CK (−75 to −67 ‰) (Fig. 2e).Compared with Treatment CK, Treatment WS was negative by 1-9 ‰ in δ 13 CH 4 -value over the season.CH 4 produced on rice roots was more de-pleted in 13 C, on average, by ∼ 16 ‰ (Treatment WS) and ∼ 4 ‰ (Treatment CK) relative to that produced in paddy soil (Fig. 2b, e) during the four rice growth stages for analysis.The δ 13 C-value of CO 2 produced in the paddy soil was relatively stable in the two treatments (∼ −18 ‰) over the four rice growth stages.On rice roots, however, it decreased obviously from the tillering and booting stages (∼ −15 ‰) to the grain-filling and ripening stages (∼ −25 ‰) (Fig. 2c, f).No significant difference was observed in δ 13 C-value of CO 2 produced either in paddy soil or on rice roots between the two treatments (P > 0.05).

CH 4 production in paddy soil and rice roots under aerobic incubation
Little CH 4 production was observed in aerobic paddy soil at the tillering, grain-filling and ripening stages, although relatively visible CH 4 production was at the booting stage (∼ 0.4 µg CH 4 g −1 d −1 ) (Fig. 3a).Significant CH 4 production on aerobic rice roots was measured at the tillering and booting stages, especially at the booting stage, reaching as  high as 5.4-22.2µg CH 4 g −1 d −1 , but little was at the grainfilling and ripening stages (Fig. 3c).CH 4 production in aerobic incubation (Fig. 3a, c) was significantly lower than in anaerobic incubation (Fig. 2a, d

CH 4 oxidation potential in paddy soil and rice roots under aerobic incubation
A similar pattern in variation of the CH 4 oxidation potential in paddy soil was observed for both treatments during the four rice growth stages (Fig. 4a), showing a relatively high beginning (4.7-8.6 µg CH 4 g −1 d −1 ) at the tillering stage, and a peak (8.1-10.9µg CH 4 g −1 d −1 ) at the grain-filling stage.However, CH 4 oxidation was very important only at the tillering stage (over 600 µg CH 4 g −1 d −1 ) on rice roots, and weakened towards the ripening stage (∼ 200 µg CH 4 g −1 d −1 ) (Fig. 4b).Compared with Treatment CK, Treatment WS showed higher CH 4 oxidation potentials in both paddy soil and rice roots at all four rice growth stages (Fig. 4a, b).

δ 13 C-value of CH 4 emitted before and after cutting of the plants
As shown in Table 2, the average δ 13 C-value of CH 4 emitted before cutting of the plants in Treatment CK was −55 ‰, being slightly higher than that in Treatment WS (−58 ‰).
Similarly, the mean δ 13 C-value of CH 4 emitted after cutting of the plants in Treatment CK (−43 ‰) was more positive than that in Treatment WS (−50 ‰).Compared with the δ 13 C-value of CH 4 emitted before cutting of the plants in both treatments, the value after cutting of the plants was significantly more positive, especially in Treatment CK (Table 2).Collectively, the latter was 8.0 ‰ in Treatment WS and 12.0 ‰ in Treatment CK higher the former (Table 2).

Organic carbon in soil and plant samples
Contents of organic carbon in soil increased slightly towards the end of the rice season, being 1.69 % and 1.85 % in Treatment CK, and 1.73 % and 1.89 % in Treatment WS at the tillering and ripening stages, respectively.However, δ 13 Cvalue of soil carbon remained stable over the season, being −7.6 ‰ and −27.8 ‰ in Treatment CK, and −27.3 ‰ and −28.0 ‰ in Treatment WS at the tillering and ripening stages, respectively.Organic carbon in plant samples was slightly lighter, with δ 13 C-value being −28.9 ‰, −29.3 ‰ and −28.6 ‰ at the tillering, booting and ripening stages, respectively.It also showed little change throughout the whole season.

Biomass and rice grain yield
Dry matter accumulations of rice plants (aboveground and underground parts) in the two treatments are presented in Table 3.They increased obviously from the tillering stage to the ripening stage and tended to be higher in Treatment CK than in Treatment WS at all four growth stages.Grain yield did not differ much between Treatment WS and Treatment CK (P >0.05) though it also tended to be higher in Treatment CK (8.89 t ha −1 ) than in Treatment WS (8.63 t ha −1 ).

Effect of straw incorporation on CH 4 production
CH 4 is an end product of methanogenic bacteria acting on methane-producing substrates under strict anaerobic conditions.Sufficient substrates and a favorable habitat for growth of methanogenic bacteria are prerequisites for CH 4 generation (Conrad, 2007).Straw incorporation provides methanogenic bacteria with abundant methane-producing substrates.Meanwhile, flooding accelerates decomposition of straw and fall in soil redox potential (Eh) (Fig. 1b), creating a favorable environment for growth of methanogens, which in turn promotes CH 4 production in the paddy field (Fig. 2a, d).Shangguan et al. (1993) found that CH 4 production rate in a straw-incorporated paddy soil, collected from a double rice cropping field in Hunan, was 1-2 times higher than that in the soil without straw incorporated.Moreover, they demonstrated that straw incorporation would be an important cause for the appearance of the first CH 4 production peak just ∼ 20 days after the field was flooded.The findings in this study also show that straw incorporation increased CH 4 production remarkably in paddy soil (Fig. 2a).Furthermore, fresh rice roots produced considerable CH 4 in an extremely anaerobic environment, showing a variation pattern similar to that in paddy soil (Fig. 2d).Previous studies have demonstrated that excised fresh rice roots themselves could produce CH 4 (Frenzel and Bosse, 1996;Conrad and Klose, 1999;Lehmann-Richter et al., 1999;Krüger et al., 2001Krüger et al., , 2002;;Zhang et al., 2011c).Straw incorporation accelerated the formation of an extremely anaerobic condition in the soil and hence the decline of soil Eh (Fig. 1b), which probably increased the population and activity of methanogens on the surface of rice roots.As a result, CH 4 production was increased (Fig. 2d).
In field conditions, variation of CH 4 concentration in soil solution reflects changes in CH 4 production of the paddy field to some extent.Under continuous flooding, CH 4 concentration in soil solution generally increased with the extension of the flooding time, and continued until the field www.biogeosciences.net/10/3375/2013/Biogeosciences, 10, 3375-3389, 2013 was drained up before the crop was harvested (Cai et al., 2009).Under intermittent irrigation in this experiment, however, it was very high at the tillering stage (before mid-season aeration) and at the booting stage (the first re-flooding period) while very low at the grain-filling and ripening stages (Fig. 1a).On the other hand, straw incorporation profoundly increased CH 4 concentration in soil solution at the tillering and booting stages (Fig. 1a, P < 0.05).This is because considerable CH 4 production in paddy soil and on rice roots was observed at the tillering and booting stages and straw incorporation highly increased their CH 4 production potentials at those periods.However, a little was measured at the grainfilling and ripening stages (Fig. 2a, d).Alberto et al. (1996) measured concentration of the CH 4 entrapped in paddy soil in the Philippines, and results also showed that it was significantly higher in plots incorporated with green manure and straw than in plots applied with chemical fertilizer.The δ 13 C-value of organic carbon was relatively lower in straw samples (−28.3 ‰) than in paddy soil (−27.4 ‰), but slightly higher than in rice roots (−28.9 ‰).In addition, it was very stable over the season both in paddy soil and rice roots.Therefore, straw per se would be neither a 13 C-enriched source for methanogenesis in paddy soil nor a 13 C-depleted source for methanogenesis on rice roots for the whole season.As a consequence, incorporation of straw increased δ 13 C-value of the CH 4 produced in paddy soil but decreased that of the CH 4 derived from rice roots (Fig. 2b, e), which is probably attributed to its different effects on pathways of CH 4 production in paddy soil and rice roots.Microbial CH 4 production per se is a process that exhibits completely different fractionation factors depending on the pathway of methanogenesis (Games et al., 1978;Gelwicks et al., 1994;Krüger et al., 2002).In paddy fields, total CH 4 production is mainly done through acetate fermentation and H 2 / CO 2 reduction (Conrad, 1999), and acetate-dependent methanogenesis is more 13 C-enriched than CO 2 -dependent methanogenesis (Sugimoto and Wada, 1993;Whiticar, 1999).In the processes of acetate and H 2 / CO 2 producing CH 4 , fractionation factors were defined by Hayes (1993): where δ 13 CH 4 (acetate) and δ 13 CH 4 (H 2 /CO 2 ) are δ 13 C-values of the CH 4 produced from acetate and H 2 /CO 2 , and δ 13 C acetate is δ 13 C-value of the acetate.
Acetate is an important intermediate in anaerobic degradation of organic matter, which is relatively stable in δ 13 Cvalues in the form of soil and plant organic carbon during the whole season (Conrad et al., 2002;Krüger et al., 2002).Generally, methyl carbon of the acetate is thought to be converted into CH 4 (Krzycki et al., 1982;Conrad et al., 2002), and an isotope fractionation factor of ε acetate/CH 4 = −21 ‰ was measured for the transformation of acetate methyl carbon into CH 4 (Gelwicks et al., 1994).Based on that, Krüger et al. (2002) estimated a measurement of δ 13 CH 4 (acetate) between −43 ‰ and −37 ‰ over the season with δ 13 C-value of the acetate (−22 to −16 ‰) extracted from the soil pore water of an Italian rice field.Therefore, δ 13 CH 4 (acetate) = −43 ‰ and −37 ‰ was assumed in this study, mainly because the δ 13 C-value of soil organic carbon (−27.4 ‰) is close to what Krüger et al. (2002) reported (−26.7 ‰).The δ 13 CH 4 (acetate) ranging from −43 ‰ to −37 ‰ is also in good agreement with those used earlier (Sugimoto and Wada, 1993;Tyler et al., 1997;Bilek et al., 1999;Conrad et al., 2002).To our knowledge, measurements of fractionation factor ε acetate/CH 4 during acetoclastic methanogenesis are still scarce (Conrad, 2005).Moreover, δ 13 C acetate changes with the growth of rice plants (Krüger et al., 2002) or the temperature of incubations (Fey et al., 2004), and it may be different from that of rice roots (Conrad et al., 2002).The constant values (−43 ‰ and −37 ‰) for δ 13 CH 4 (acetate) were applied both in paddy soil and on rice roots in this study due to the lack of measurements of ε acetate/CH 4 and δ 13 C acetate .
It is also essential to set a fractionation factor α CO 2 /CH 4 that occurs during H 2 /CO 2 reduction to CH 4 for this experiment.Previous studies have showed that hydrogenotrophic methanogenesis expresses a stronger kinetic isotope effect than acetoclastic methanogenesis does (Whiticar, 1999;Conrad, 2005).By summarizing the reported measurements, Zhang et al. (2009) showed that α CO 2 /CH 4 was in the range of 1.025-1.083.Fey et al. (2004) found α CO 2 /CH 4 in flooded paddy soil decreased with rising temperature in anaerobic incubation, with value of 1.083 at 10 • C, 1.079 at 25 • C, and 1.073 at 37 • C. Since the incubation in this study varied in the range from 17.6 • C to 29.7 • C and averaged 25.1 • C in temperature, α CO 2 /CH 4 = 1.079 was hence considered to be reasonable for calculation of δ 13 CH 4 (H 2 /CO 2 ) .Fortunately, it has been validated in the other experiments we conducted (Zhang et al., 2011b(Zhang et al., , 2012)).It is much higher than those in earlier reports (Sugimoto and Wada, 1993;Tyler et al., 1997;Bilek et al., 1999;Chidthaisong et al., 2002;Krüger et al., 2002;Valentine et al., 2004), indicating fractionation factor α CO 2 /CH 4 should not be assumed to be the same for different experiments and situations.Such differences may result from variation in the community structure of methanogens with the soil and the duration and condition of incubation (Chin et al., 1999;Lueders and Friedrich, 2000;Chidthaisong et al., 2002;Conrad et al., 2002;Krüger et al., 2002;Ramakrishnan et al., 2002).
When δ 13 CH 4 (acetate) , δ 13 CH 4 (H 2 /CO 2 ) , and δ 13 C-value of CH 4 (δ 13 CH 4 ) produced in paddy soil and rice roots were obtained, relative contribution of acetate to total CH 4 production (F ac ) could be estimated in line with Eq. (4) (Tyler   et al., 1997;Bilek et al., 1999;Krüger et al., 2002): As shown in Table 4, hydrogenotrophic methanogenesis in paddy soil was very important at the tillering and booting stages (∼ 50-70 %) while acetoclastic methanogenesis dominated at the grain-filling and ripening stages (∼ 60-90 %).Similarly, Krüger et al. (2002) also found that acetatedependent methanogenesis dominated the end of the season, whereas H 2 / CO 2 -dependent methanogenesis was very important at the beginning of the season.Acetoclastic methanogenesis was more important in Treatment WS than in Treatment CK at all four rice growth stages, with F ac -value higher by ∼ 10-30 %.It shows that incorporation of straw supplies abundant substrates for soil CH 4 production, thus promoting acetate-dependent methanogenesis.Wang et al. (1995) found that application of the organic fertilizers significantly increases the content of soil organic acid, which is positively related to the rate of CH 4 emission from rice field.Unfortunately, contents of the organic acids, especially acetate, in soil solution were not measured simultaneously.The pathways of methanogenesis in paddy soil have been considerably observed in Italy and Japan, and they are similar to the measurements in this study (Table 5).
For rice roots, however, hydrogenotrophic methanogenesis was dominant (∼ 50-90 %) in both treatments at all four rice growth stages and more important in Treatment WS than in Treatment CK, with a mean F ac -value being ∼ 5-20 % lower (Table 6).Additionally, the average F ac -value was 6 % higher (Treatment CK) and 30 % higher (Treatment WS) in paddy soil than on rice roots (Table 6).A similarly high contribution of hydrogenotrophic methanogenesis to total CH 4 production on rice roots was observed in radiotracer experiments (Conrad et al., 2000(Conrad et al., , 2002)).Measurements of stable carbon isotope also showed that methanogenesis on excised rice roots was mostly from H 2 / CO 2 throughout the season (Krüger et al., 2002).The methanogens population on rice roots is different from that of surrounding soil (Grosskopf et al., 1998a, b;Lehmann-Richter et al., 1999), which may be the reason for the high hydrogenotrophic methanogenesis.Previous studies demonstrated that CH 4 was predominantly produced from H 2 / CO 2 on rice roots by Rice Cluster I (RC-I) methanogens (Lehmann-Richter et al., 1999;Lueders et al., 2001;Lu et al., 2005;Lu and Conrad, 2005;Conrad et al., 2006).Therefore, straw incorporation would accelerate the growth and activity of RC-I methanogens, hence promoting CO 2 -dependent methanogenesis relative to Treatment CK (Table 6).Organic carbon being slightly lighter in plant samples than in soil samples might be another possible reason for F ac -value being lower in paddy soil than on rice roots (Tables 4 and 6).Compared with previous measurements, the relative contribution of acetate to total methanogenesis on fresh roots in this study appeared to be slightly lower, particularly in Treatment WS (Table 5).

Effect of straw incorporation on CH 4 oxidation
Straw incorporation not only provides sufficient substrates for methanogens in paddy soil, thus promoting methanogenesis directly, but also affects soil CH 4 oxidizing capacity indirectly.Straw incorporation affects CH 4 oxidation mainly through its influence on methanotrophic population and activity.Previous studies have shown that a high concentration of CH 4 stimulates growth of methanotrophs and their activity in oxidization (Bender and Conrad, 1995;Arif et al., 1996).As shown in Figs.2a and 4a, it increased the CH 4 production capacity in paddy soil, and as a consequence, the CH 4 oxidizing ability, as well.On the other hand, it is not only the oxygen secreted from the roots that oxidize CH 4 in the rhizosphere (Butterbach et al., 1997), but also the roots per se that have a strong CH 4 oxidization capacity (Bosse and Frenzel, 1997;Dan et al., 2001).Krüger et al. (2002) found that CH 4 oxidation rates on excised fresh rice roots were highest at the beginning of the season and then declined, which is in agreement with the measurements in this study.Straw incorporation enhanced CH 4 oxidation potential on rice roots to some extent (Fig. 4b), which is possibly attributed to the growth and activity of methanotrophs on the surface of rice roots stimulated by straw decomposition.Unfortunately, community structure of the methanotrophs in the soil was not analyzed in this study, and therefore further research is needed in this aspect.
Besides the direct in vitro measurements of CH 4 oxidation above, δ 13 C-values of the CH 4 from various compartments of the rice field were applied to estimation of the fraction of CH 4 oxidized (F ox ) by using the following steady-state mass balance equation (Stevens and Engelkemeir, 1988;Tyler et al., 1997): (1/α ox − 1) × δ 13 CH 4 (oxidized) + 1000 , where α ox stands for fractionation factor during CH 4 oxidation, δ 13 CH 4 (original) for δ 13 C-value of the initial pool of CH 4 that is produced in soil under aerobic incubation (Fig. 2a), and δ 13 CH 4 (oxidized) is the δ 13 C-value of remaining unoxidized CH 4 , generally estimated from the measurements of δ 13 CH 4 (emitted) corrected with plant transport fractionation (Tyler et al., 1997;Krüger et al., 2002): where δ 13 CH 4 (emitted) stands for δ 13 C-value of CH 4 emitted from the rice field, and ε transport for transport fractionation factor, with a range of −12.0 to −8.0 ‰ in the present study (for detailed description, please see Sect.4.3.below).
The fraction of CH 4 oxidized (F ox ) was calculated using Eq. ( 5) based on the values of δ 13 CH 4 (original) , δ 13 CH 4 (oxidized) and α ox (1.025 in Treatment CK and 1.033 in Treatment WS) subsequently referred to (for detailed description, please see Sect.4.3.below) and shown in Table 7.It peaked at the tillering stage (60-101 %), declined gradually and reached the lowest (−19-45 %) at the ripening stage both in Treatment CK and Treatment WS (Table 7).This was well in agreement with the findings of Krüger et al. (2001Krüger et al. ( , 2002)), who reported that the value of F ox was the highest at the beginning of the season, turned lower and lower towards the end of the season, and even dropped below zero as depicted in the present study at last (Table 7).Contrary to its effect on CH 4 oxidation potential of paddy soil, straw incorporation reduced the value of F ox , in terms of percentage, by 41-71 % during the four rice growth stages (Table 7).A possible explanation was that straw incorporation increased both CH 4 production and oxidation potentials, particularly the former, to a larger extent (Figs.2a and 4a), which eventually caused a decrease in F ox -value (Table 7).It was more visible when the situations of high CH 4 production potential and low CH 4 oxidation potential appeared simultaneously at the booting stage (Figs.2a and 4a, Table 7), with F ox -value of 71 % lower in Treatment WS than in Treatment CK.CH 4 production rates in aerobic incubation were very low relative to those in anaerobic incubation (Figs. 2a,d,3a and c), which suggests that strong oxidation happens therein.On the other hand, it is quite clear in this study that CH 4 produced in aerobic incubation was much more positive than that in anoxic incubation (Figs. 2b and 3b).This further shows that CH 4 in the former has been oxidized intensively relative to that in the latter.So the δ 13 C-values of CH 4 aerobically produced in paddy soil were likely to represent δ 13 CH 4 (oxidized) .An analogous calculation was tentatively conducted with Eq. ( 5), using the stable carbon isotope technique in the present study.Similar temporal variation   of F ox -value in the two treatments was observed (Table 8), being the highest (41-101 %) at the tillering stage and the lowest at the ripening stage (−3-45 %).Moreover, F ox -value was 45-68 % lower in Treatment WS than in Treatment CK, which was consistent with the results reported before (Table 7).It indicates that the δ 13 CH 4 (oxidized) obtained in this way, to some extent, may be used to represent the δ 13 C-value of CH 4 that remains after being oxidized but has not yet been emitted to the atmosphere if it is hard to measure ε transport in rice fields.
Compared with the former reports, the measurements of F ox -value in this study were significantly higher (Table 5).The differences in α ox between these studies were a possible reason, because the α ox used in this experiment was 1.025-1.033,while in others 1.038 was used to calculate F ox -value (Krüger et al., 2002;Krüger and Frenzel, 2003;Zhang et al., 2012).Therefore, α ox = 1.038 was used instead in estimating F ox in order to offset the discrepancy caused by different values of α ox .For the present study, F ox -values were still reduced by ∼ 20-40 % in Treatment WS relative to that in Treatment CK (Tables 7 and 8), further suggesting that straw incorporation is an important factor, instead of α ox , influencing the measurement of F ox -value.More in-terestingly, it was still much higher than the measurements in previous studies in Italy (Krüger et al., 2002;Krüger and Frenzel, 2003), especially in Treatment CK (Tables 7 and  8).However, it was similar to the data reported by Zhang et al. (2012), who found F ox -value was ∼ 20-70 % in continuous flooding plots and ∼ 50-80 % in intermittent irrigation plots.This indicates that the fields under the special water management in China, i.e., intermittent irrigation (this study and Zhang et al., 2012), would increase CH 4 oxidation in comparison with those under continuous flooding (Krüger and Frenzel, 2003) or those that had just a brief period of drainage (Krüger et al., 2002).Therefore, α ox itself in the present study may not be a key factor influencing F ox -value relative to early reports in different conditions (Tyler et al., 1997;Bilek et al., 1999;Krüger et al., 2002;Krüger and Frenzel, 2003;Conrad et al., 2005).

Effect of straw incorporation on carbon isotope fractionation during CH 4 oxidation and transport
When the stable carbon isotope method is used for calculating F ox , oxidation fractionation factor (α ox ) has to be taken into consideration (Tyler et al., 1997;Krüger et   a Calculated with Eq. ( 5) using the δ 13 C-values of CH 4 anaerobically produced in paddy soil (Fig. 2b) for δ 13 CH 4 (original) and the δ 13 C-values of CH 4 aerobically produced in paddy soil (Fig. 3b) for δ 13 CH 4 (oxidized) .
2002).The oxidation fractionation is caused by methanotrophs.In the closed-system incubation, fractionation factor α ox is known to be calculated according to the Rayleigh equation (Coleman et al., 1981;Liptay et al., 1998): − log δ 13 CH 4 (final) + 1000 / log f, where δ 13 CH 4 (initial) stands for δ 13 C-value of CH 4 at time 0, δ 13 CH 4 (final) for δ 13 C-value of CH 4 at time t, and f (%) for percentage of CH 4 remaining at time t.
In paddy soil 28.3 • C in temperature, α ox = 1.025 in Treatment CK and 1.033 in Treatment WS was observed in this study, which was well in agreement with the measurements before (1.025-1.038,Coleman et al., 1981;Chanton and Liptay, 2000).To our knowledge, α ox was firstly measured in methanotroph-enriched cultures (Coleman et al., 1981) and then mainly in landfill cover soils (Liptay et al., 1998;Chanton et al., 1999;Chanton and Liptay, 2000;Mahieu et al., 2006;Chanton et al., 2008a, b).Although nothing was known about α ox in paddy soil before, it (1.025-1.038)was adopted considerably in paddy field experiments (Tyler et al., 1997;Bilek et al., 1999;Krüger et al., 2002;Krüger and Frenzel, 2003;Conrad and Klose, 2005).It is affected by temperature (Chanton and Liptay, 2000;Chanton et al., 2008a), methanotrophs (Coleman et al., 1981;Krüger et al., 2002), and soils (Tyler et al., 1994;Chanton and Liptay, 2000;Snover and Quay, 2000).The differences in α ox between Treatment CK and Treatment WS might be attributed to methanotrophic bacteria in the soil.Methanotrophs preferentially oxidize 12 CH 4 , leaving the residual CH 4 13 Cenriched (Whiticar, 1999;Venkiteswaran et al., 2005), which results in a shift in the isotopic fractionation.There is such a possibility that the higher the population and activity of methanotrophs, the more the 12 CH 4 being preferentially consumed.Subsequently, the more the residual CH 4 enriched in 13 C, the bigger the fractionation after CH 4 oxidation.Therefore, straw incorporation increased α ox , which is probably ascribed to its promotion of CH 4 oxidation by stimulating methanotrophic bacteria in Treatment WS relative to Treatment CK (Fig. 4a).Although it has been reported consider-ably, the lack of knowledge on α ox in paddy soil calls for more efforts in the further study.
In rice fields, most unoxidized CH 4 escapes into the atmosphere through the aerenchyma of the rice plants.In the process of CH 4 transport via plants, significant transport fractionation is observed (Chanton, 2005).During the rice season, the CH 4 transport fractionation factor ε transport is known to be equivalent to the difference between δ 13 C-values of the emitted and aerenchymatic CH 4 (Tyler et al., 1997;Bilek et al., 1999;Krüger et al., 2002).A value of ε transport was −8.0 ‰ in Treatment WS and −12.0 ‰ in Treatment CK estimated accordingly from the data of Table 2. Similar differences (∼ −12 ‰) were observed in former reports (Chanton et al., 1997;Tyler et al., 1997;Bilek et al., 1999).It is a matter of fact, in general, that both CH 4 transport efficiency (Jia et al., 2002) and value of ε transport (Krüger et al., 2002;Conrad and Klose, 2005) are significantly affected by the growth of plants during the rice season.The biomass of rice plants including both aboveground and underground parts at all four rice growth stages was lower in Treatment WS than in Treatment CK (Table 3), which suggests that the growth of rice crop has been controlled by straw incorporation.On the other hand, CH 4 transportation by plants is basically a diffusion process, and a small difference in condition may bring about a great difference in isotopic composition (Chanton, 2005).Pressure and partial pressure in the rhizosphere is possibly higher in Treatment WS than in Treatment CK as well as the δ 13 C-value of CH 4 in the soil.Therefore, those differences in physical condition may cause differences in transport fractionation.Although the processes resulting in the difference in ε transport are not fully understood, the differences in physical conditions, such as growth of rice plants and pressure in the rhizosphere, are likely to affect the diffusion process and consequently the ε transport .

Conclusions
The study on the mechanism of CH 4 emission from a Chinese rice field demonstrated that straw incorporation obviously increased CH 4 production and oxidation potentials in Biogeosciences, 10, 3375-3389, 2013 www.biogeosciences.net/10/3375/2013/paddy soil and on rice roots.What is more, the effect of straw incorporation on methanogenic pathways and fraction of CH 4 oxidized was quantified by measuring stable carbon isotopic signatures.The results show that straw incorporation increased the contribution of acetate to methanogenesis in paddy soil by ∼ 10-30 %, but decreased on rice roots by ∼ 5-20 %, and acetoclastic methanogenesis was more important in the former than in the latter.Furthermore, straw incorporation significantly decreased the fraction of CH 4 oxidized by 41-71 %, which is likely ascribed to the fact that CH 4 production potential was increased to a larger extent than CH 4 oxidation potential was.This indicates that the production of CH 4 increased directly by straw incorporation will indirectly reduce the fraction of CH 4 oxidized, thus significantly decreasing CH 4 emission from the rice fields.These findings are beneficial to enhance the understandings of the effect of straw incorporation on CH 4 emission and the mechanism at the soil micro-process level and further to supply scientific advice and available options for mitigating CH 4 emissions from rice-based ecosystems.Although it may be difficult to estimate its quantitative effects on CH 4 production and oxidation on a regional or a global scale, the variation pattern of methanogenesis being increased while the fraction of CH 4 oxidized being reduced by straw incorporation is probably not changed.More relevant investigations in different conditions should be done in the future.This study also contributes another important question related to carbon isotope fractionation during the processes of CH 4 oxidation and transport.Although it is very limited, to our knowledge so far, in paddy soil, the fractionation factor α ox varied in the range of 1.025-1.038,which was well in agreement with previous reports on different soils.What is more, it was increased by straw incorporation, which probably resulted from the effects of straw incorporation stimulating methanotrophs and hence promoting CH 4 oxidation in the soil.Difference between δ 13 C-values of the emitted and aerenchymatic CH 4 indicates a fractionation factor ε transport of −12.0 ‰ to −8.0 ‰, which was to some extent controlled by straw incorporation.The reason is likely that the diffusion process in transport is affected by growth of rice plants and pressure in the rhizosphere, though the processes causing the difference are not clearly known.As a traditional agricultural management, straw incorporation is estimated to be the most sensitive factor influencing CH 4 emission from rice fields.Produced CH 4 partially oxidized by the paddy fields has a very important effect on regulating the atmospheric CH 4 .By quantifying the fractionation factors (α ox and ε transport ), one can not only better understand the kinetic isotope effect in the processes of CH 4 emission but also more accurately estimate CH 4 oxidation, which will help to put constraints on global CH 4 sinks.Therefore, it is essential to contribute more efforts to the study on methanotrophic microbial communities and fractionation factors α ox and ε transport to better elucidate the processes of CH 4 emission from rice fields as affected by straw incorporation.
by Copernicus Publications on behalf of the European Geosciences Union.G. B. Zhang et al.: Pathway of CH 4 production, fraction of CH 4 oxidized, and 13 C isotope fractionation

Table 1 .
Schedule of fertilizer application and water management during the rice season.

Table 2
. δ 13 C-values of CH 4 (‰) emitted before and after cutting of the plants.WS: straw incorporation, CK: without straw incorporation.

Table 5 .
Overview of the contribution of acetate-dependent CH 4 production (F ac ) and the fraction of CH 4 oxidized (F ox ) in paddy fields calculated with Eqs.(4) and (5).
CF the field was under continuous flooding, II the field was under intermittent irrigation.

Table 7 .
Fraction of CH 4 oxidized (F ox ) a in paddy fields at the four rice growth stages; mean ± SD, n = 3. WS: straw incorporation, CK: without straw incorporation.
a Calculated with Eq.

Table 8 .
Fraction of CH 4 oxidized (F ox ) a in paddy fields at the four rice growth stages; mean ± SD, n = 3. WS: straw incorporation, CK: without straw incorporation.