Mitigation of Greenhouse Gas Emissions by Water Management in a Forage Rice Paddy Field Supplemented with Dry-Thermophilic Anaerobic Digestion Residue

Mitigation of Greenhouse Gas Emissions by WaterManagement in a Forage Rice Paddy Field Supplementedwith Dry-Thermophilic Anaerobic Digestion Residue

S. Riya & M. Katayama & E. Takahashi & S. Zhou &

A. Terada & M. Hosomi

Received: 15 April 2014 /Accepted: 7 August 2014 /Published online: 22 August 2014# Springer International Publishing Switzerland 2014

Abstract Dry-thermophilic anaerobic co-digestion(DTAD) can be used to treat forage rice straw and pigmanure and generate biogas as an energy source. Solidresidue produced from DTAD process can be used as afertilizer in forage rice fields, while addition of theresidue could increase methane (CH4) and nitrous oxide(N2O) emissions from the soil. We evaluated the effectsof adding DTAD residue and water management onCH4 and N2O emissions from a forage rice field. Threetreatments were evaluated: (a) 100 kg N·ha−1 chemicalfertilizer and continuous flooding (CC); (b) residue ad-dition (300 kg N·ha−1 DTAD residue) with continuousflooding (RC); and (c) residue addition with intermittentirrigation (RI). RC and RI showed higher CH4 fluxesthan CC throughout the growing period. After a mid-summer drainage, RI showed higher soil Eh values andlower CH4 fluxes (mean, 7.6 mg C·m−2·h−1) than thosein RC (mean, 18.6 mg C·m−2·h−1). Abundance ofmcrAgene copy number was not different between RC andRI, suggesting CH4 flux was reduced by suppression ofmethanogenic activity by intermittent irrigation.

Cumulative CH4 emissions during the cultivationperiod were 105, 509, and 306 kg C·ha−1 in CC,RC, and RI, respectively. N2O fluxes were withindetection limits in all treatments. Our results, to ourknowledge, are the first to show greenhouse gasemission from forage rice fields supplemented withDTAD residue and of the effectiveness of watermanagement in CH4 mitigation.

Keywords Methane . Forage rice .Water management .

Methanogenic archaea . Dry-thermophilic anaerobicdigestion . Biogas residue

1 Introduction

Biogas production from agricultural waste is an effec-tive strategy for waste management because it producesenergy and mitigates greenhouse gas emissions(Arthurson 2009). Among the biogas productionmethods, dry anaerobic digestion has received greatattention in recent years. It is generally carried out witha higher solids content (>10 %) than that used in wetanaerobic digestion, and it treats larger volumes ofwastes (Jha et al. 2011). Because dry thermophilic an-aerobic digestion (DTAD) process generates a solidresidue, beneficial use of the residue is important interms of material cycling. In a previous study, Zhouet al. (2013) conducted experiments on DTAD of swinemanure (pig dung+urine) combined with forage ricestraw and showed that stable fermentation occurred inthis system. If this residue can be used in forage rice

Water Air Soil Pollut (2014) 225:2118DOI 10.1007/s11270-014-2118-3

S. Riya (*) :M. Katayama : E. Takahashi :A. Terada :M. HosomiDepartment of Chemical Engineering, Faculty ofEngineering, Tokyo University of Agriculture andTechnology,2-24-16 Naka-cho, Koganei, Tokyo 184-8588, Japane-mail: [email protected]

S. ZhouEco-environmental Protection Research Institute, ShanghaiAcademy of Agricultural Sciences,1000 Jinqi Road, Fengxian, Shanghai 201403, China

fields as fertilizer, harvested rice can be used as pig feed.The remaining leaves and stems can be treated in DTADas organic carbon source. Thus, DTAD and forage riceproduction could represent a sustainable pig manuretreatment system.

Rice fields are a major contributor to methane (CH4)emissions. According to estimates by the United StatesEnvironmental Protection Agency (US-EPA), CH4

emissions from rice paddy fields account for approxi-mately 7 % of anthropogenic global CH4 emissions(US-EPA 2012). In rice soil, CH4 is produced by me-thanogenic archaea from either H2/CO2 or acetate,which are products of anaerobic decomposition of or-ganic matter (Le Mer and Roger 2001). Therefore,DTAD residue application in forage rice fields wouldprovide organic carbon for growth of methanogenicarchaea.

Forage rice needs more nitrogen than food rice tosupport high biomass production (Kyaw et al. 2005).Because whole crop is utilized as a feed for pig orcattle, returning pig or cattle waste is important fornutrient management in forage rice cultivation. Ithas been reported that more applications of organicwaste increased biomass as well as greenhouse gasemission from forage rice fields (Sasada et al. 2011;Riya et al. 2012; Win et al. 2010). Sasada et al.(2011) found that application of anaerobicallydigested cattle/pig slurry caused higher CH4 emis-sion than that by the application of a chemicalfertilizer. On the other hand, Riya et al. (2012)observed a rapid increase in CH4 fluxes after top-dressing with liquid cattle waste. While liquid or-ganic waste contained mineral nitrogen and labilecarbon, DTAD residue contained more organic ni-trogen and carbon. In addition, the residue should beapplied before transplanting, at one time, as straw orcompost incorporation. Therefore, DTAD applica-tion in forage rice paddy fields may differently af-fect soil physicochemical properties and exhibit adifferent pattern and intensity of CH4 emission fromorganic liquid waste application. However, no infor-mation about relationship between dynamics of CH4

flux and environmental factors in a forage rice fieldincorporated with DTAD residue is available.

Water management is important for rice produc-tion and also has the potential to mitigate CH4 emis-sions (Yagi et al. 1996; Zou et al. 2005). In a ricefield managed by midsummer drainage and

intermittent irrigation, CH4 emissions during culti-vation period are mitigated by 65 % and 36 % withand without wheat straw incorporation, respectively(Zou et al. 2005). Recent review determined thatCH4 emission in continuously flooded rice fields isas much as 90 % higher compared with other watermanagement methods, independent from straw in-corporation (Sanchis et al. 2012). Thus, water man-agement is effective to mitigate CH4 emission in arice field amended with organic matter. Althoughthere have been many reports about effectiveness ofwater management on CH4 emission, contrastingresults have been reported about response of methan-ogenic archaea to water management. For example,Jiao et al. (2006) reported lower number ofmethanogens in a paddy soil managed by intermit-tent irrigation than in those managed by continuousflooding, while Watanabe et al. (2013) showed noclear reduction of methanogenic archaea. Therefore,the effect of water management on CH4 mitigation aswell as abundance of methanogenic archaea shouldbe evaluated in a forage rice field applied withDTAD residue.

Riya et al. (2012) found a rapid increase in ni-trous oxide (N2O) emission, a potent greenhouse gasthat has 298-times higher global warming potentialthan that by CO2 (IPCC 2007), during drainage afteran application of liquid cattle manure. N2O emissionduring drainage is widely reported for rice fields(Ma et al. 2007; Zou et al. 2005). This impact ontotal greenhouse gas emission is emphasized espe-cially for high N-fertilized rice fields (Riya et al.2012). According to our system analysis, forage ricefields should receive about 300 kg N·ha−1 of DTADresidue for production of forage for pig and utiliza-tion of the residue (see Subsection 2.1). This nitro-gen rate is about three times higher than theJapanese conventional nitrogen fertilization rate(100 kg N ha−1; Nishimura et al. 2004). Therefore,the combined global warming impact of CH4 andN2O should be evaluated.

The objective of this study was to clarify theeffects of incorporating DTAD residue on green-house gas emissions from a forage rice field and todetermine whether water management could miti-gate these effects. Therefore, we added DTAD resi-due to a forage rice field and compared the effects oflocal water management and intermittent irrigation

2118, Page 2 of 13 Water Air Soil Pollut (2014) 225:2118

strategies on greenhouse gas emissions. To confirmthe impact of DTAD residue application and watermanagement on CH4 emission, several environmen-tal factors affecting CH4 production, functional geneof methanogenic archaea, and plant parameters werealso measured.

2 Materials and Methods

2.1 Study Site and Experimental Layout

The field study was carried out in a local farmer’srice field (1,000 m2) in the city of Namegata,Ibaraki Prefecture, Japan (36°7′ N, 140°23′ E) in2013. The soil was classified as a sandy loam soil.The total carbon and total nitrogen contents were0.90 % and 0.04 %, respectively, in the 0–5 cmlayer of soil, and 1.11 % and 0.10 %, respectively,in the 5–20 cm layer of soil.

We applied three treatments: (a) CC, 100 kg N·ha−1 chemical fertilizer (NH4

+, P2O5, and K2O, each8 %) and continuous flooding; (b) RC, 300 kg N·ha−1

DTAD residue (on a total N basis) and continuousflooding; and (c) RI, 300 kg N·ha−1 DTAD residueand intermittent irrigation. The total organic carbonin the residue was 5496 kg C·ha−1. Nitrogen fertili-zation rate in CC is based on conventional manage-ment of rice, whereas the amount of 300 kg N ha−1 inRC and RI is determined from the treatment ofDTAD residue generated from DTAD facility treatingmanure from 1,000 pigs. In the calculations, the areaof forage rice paddy field was estimated from landarea required to supply forage for 1,000 pigs. There-fore, this study examined greenhouse gas (GHG)emissions when using DTAD residue in the manage-ment of forage rice fields. In the experimental ricefield, nine plots, each 1 m×3 m, were constructed byinserting a plastic frame into the hard subsoil layer.The frame was 15 cm high above the soil surface.Each treatment was assigned three plots.

Chemical fertilizer and DTAD residue as basal fertil-izer were applied at a rate of 100 kg N ha−1 in CC and300 kg N ha−1 in RC and RI onMay 22. No topdressingwas applied during cultivation. The DTAD residue wasmixed in at 0–10 cm depth. Rice seedlings of Takanari(cultivar of Oryza sativa L.) were transplanted at a

planting density of 18 hills·m−2 onMay 29 (0 days aftertransplanting [DAT]). Midsummer drainage was con-ducted in all treatments from July 10 to 16 (42–48DAT). Intermittent irrigation was conducted from 48 to106 DAT in RI. Final drainage to dry the soil was carriedout on September 11 (106 DAT). Rice plants wereharvested on October 13 (137 DAT).

2.2 Preparation of DTAD Residue

A semi-batch digester was constructed to prepare theDTAD residue. Pig manure was co-digested withforage rice straw at 55 °C, in a 73 % water content,and for 30–40 days of sludge retention time in thedigester. The inoculum used for digestion was takenfrom a digester used for DTAD of food waste, paper,and twigs. The degradation rate of volatile solidsduring digestion was approximately 50 %. TheDTAD residue was prepared by drying fresh DTADresidue in N2 atmosphere to degrade readily decom-posable residual organic matter and to improve han-dling and storage. The total carbon and nitrogencontents of the residue were 30.3 % and 1.69 %(on a wet weight basis), respectively. The C/N ratiowas close to 18.

2.3 Measurement of CH4 and N2O Emissions

Fluxes of CH4 and N2O were evaluated using theclosed chamber method (Riya et al. 2012). Threechamber bases (each 0.40 m long, 0.20 m wide,and 0.20 m high) were inserted to a depth of approx-imately 5 cm in each plot 1 h before gas sampling.The chamber base had a water channel to make thegas chamber airtight. The chambers were equippedwith a thermometer, pressure-adjusting bag, fan, andthree-way stopcock. Gas samples were withdrawn at0, 10, and 20 min after the chambers were set ontheir bases and were transferred to pre-evacuatedglass vials (SVG-10, Nichiden-Rika Glass Co. Ltd.,Hyougo, Japan). During gas sampling, water tableand soil Eh (see later) were measured.

Methane concentration in gas samples was analyzedusing GC-14B gas chromatograph with a flame ioniza-tion detector (Shimadzu, Kyoto, Japan), and N2O con-centrations were measured using GC-14A gas chro-matograph with an electron capture detector (Shimadzu,Kyoto, Japan). The gas flux from each plot to the

Water Air Soil Pollut (2014) 225:2118 Page 3 of 13, 2118

atmosphere was calculated using Eq. (1) based on anincrease in gas concentration in the chamber over time:

F ¼ ΔC=Δt � V=V0 � T0=T � M � 1=A ð1Þwhere F is the flux of CH4 or N2O (g·m−2·h−1),

ΔC/Δt is the change in CH4 or N2O concentration inthe chamber per unit time (ppmv·h−1), V is thevolume of the chamber (m3), V0 is the molar volumeof ideal gas (0.0224 m3·mol−1), T0 is 273 K, T is theair temperature in the chamber (K), M is the molar

weight of CH4 or N2O (g ·mol−1), and A is thesectional area of the base (m2). Cumulative CH4

and N2O emissions during cultivation period wereestimated by trapezoidal integration of the mean fluxover time.

To evaluate the global warming impact of forage ricefields, we calculated the cumulative CO2-equivalent(CO2-eq) emission using the 100-year global warmingpotential (25 for CH4; 298 for N2O) (IPCC 2007), asfollows:

CO2−eq ¼ 25 � ECH4‐C � 16.12 þ 298 � En2O‐N � 44

.28

� �.1000 ð2Þ

where CO2-eq is the cumulative CO2-eq emission (tCO2·ha

−1), and ECH4–C and EN2O–N are cumulativeCH4–C and N2O–N emissions (kg·ha−1), respectively.

2.4 Measurement of Environmental Factors

To measure soil Eh at 5 cm depth, two platinumelectrodes (EP-201, Fujiwara Scientific Co. LTD.,Tokyo, Japan) were permanently installed in eachplot. Soil Eh was measured using a portable Ehmeter (PRN-41, DKK-TOA Corporation, Tokyo,Japan) with a platinum electrode connected to anAg/AgCl reference electrode (4400, DKK-TOACorporation).

At the time of rice transplanting, two porouscups (DIK-8390, Daiki Rika, Saitama, Japan) werepermanently inserted at 5 cm depth in each plot.During the experimental period, pore water wasperiodically sampled from the cups using a sy-ringe. Sampled pore water was filtered through a0.45-μm membrane filter and stored at 4 °C untilanalysis. Dissolved organic carbon (DOC) concen-trations were measured with a total organic carbonanalyzer (TOC 5000A; Shimadzu). Concentrationsof NH4

+ and NO3− were analyzed by using ion

chromatography (ICS-90 for NH4+ and ICS-1000

for NO3−; Dionex, Sunnyvale, CA, USA).

2.5 Analysis of mcrA Copy Number in the Soil

Three soil cores (at 0–5 cm depth) were collected on 34,79, 113, and 137 DAT from each plot for molecular

biological analyses and stored at 4 °C until DNA ex-traction. Total DNA was extracted from soil samplesusing Extrap Soil DNA Kit Plus ver. 2 (Nippon Steel& Sumikin Eco-Tech Corporation, Tokyo, Japan) ac-cording to the manufacturer’s protocol. Real-time PCRwas carried out using CFX96 real-time PCR system(Bio-Rad, Hercules, CA, USA) to determine the mcrAcopy number using primers mlas and mcrA-rev(Steinberg and Regan 2009). Each reaction was per-formed in 20 μL aliquots containing 10 μL Sso FastTM

EvaGreen Supermix (Bio-Rad), 0.5 μL of each forwardand reverse primer, 5.0 μL DNA sample, and 4.0 μLMilliQ water. The conditions for real-time PCR were asfollows: preheating at 94 °C for 300 s, followed by40 cycles of denaturation at 94 °C for 25 s, annealingat 65 °C for 30 s, and elongation at 72 °C for 45 s. Amelting curve analysis was conducted by increasing thetemperature from 65 °C to 95 °C at 0.5 °C incrementsevery 5 s. Measurements were performed in triplicate.

2.6 Rice Plants

A single rice plant including root was randomlytaken from each plot at 34, 79, and 113 DAT tomeasure aboveground biomass and root biomass.At harvest (137 DAT), aboveground biomass of18 rice plants were harvested from each plot todetermine the yield of rice plants. One of the 18rice plants was randomly selected as a rice plant tobe used for growth parameter determination. Weightof the aboveground and root biomass was deter-mined after air-drying.

2118, Page 4 of 13 Water Air Soil Pollut (2014) 225:2118

2.7 Statistical Analysis

One-way analysis of variance (ANOVA) was performedto analyze the variation in greenhouse gas emissions andyield of harvested rice plants among treatments. Signif-icant difference between different treatments and be-tween different days of treatment was analyzed formcrAcopy number and plant growth parameters, respectively,by least significant difference test. Correlation analyseswere carried out between DOC concentration and CH4

flux for a given period. All statistical analyses wereperformed using SPSS v.16.0 (SPSS Inc., USA).

3 Results

3.1 Methane and Nitrous Oxide Fluxes

In CC treatment, CH4 flux gradually increased from 0.5to 5.8 mg C·m−2·h−1 before the midsummer drainage(Fig. 1). CH4 fluxes in RC and RI were greater thanthose in CC, fluctuating between 6.5 and 42.3 mg C·m−2·h−1. However, during midsummer drainage, CH4

fluxes decreased to 0.3–2.2 mg C m−2·h−1. From 48 to106 DAT, CH4 flux in CC again gradually increased.CH4 fluxes in RC and RI increased more rapidly thanthose in CC. The increase in CH4 flux in RI was smallerthan that in RC from 48 to 106 DAT.

During the entire cultivation period, N2O fluxes fluc-tuated between −24.7 and 13.9 μg N m−2·h−1 (data notshown). These values were within the detection limit(58 μg N m−2·h−1). Therefore, N2O fluxes were consid-ered as zero throughout the cultivation period.

3.2 Cumulative Greenhouse Gas Emissions

In CC, the cumulative CH4 emission was 105 kg C ha−1,significantly lower than in RC (509 kg C ha−1; Table 1).The cumulative CH4 emission in RI was 306 kg C ha−1,approximately 60 % of that in RC.

Table 1 also shows CO2-eq for each treatment, cal-culated from CH4 emission values and the 100-yearglobal warming potential factor. N2O emissions werenot included in the calculations, as mentioned above.The lowest CO2-eq was 3.5 t CO2 ha

−1 in CC. CO2-eqvalues in RC and RI were 17.0 and 10.2 t CO2 ha

−1,respectively. The fluctuation in CO2-eq values among

the treatments had the same trend as the one in CH4

emissions.

3.3 Environmental Factors

The changes in DOC and NH4–N concentrations in thepore water at 5 cm soil depths are shown in Fig. 2a, b,respectively. In CC, DOC concentration at 5 cm depthgradually decreased from 23 mg C L−1 (7 DAT) to9.2 mg C L−1 (43 DAT); then, after the midsummerdrainage, it gradually increased to 23 mg C L−1 (97DAT). In RC and RI, DOC concentrations were higherthan those in CC. At the first sampling time (7 DAT),DOC concentration was 150 mg L−1 in RC and RI, sixtimes higher than in CC, and it decreased over time.DOC at 5 cm soil depth in RC and RI sharply decreasedduring the midsummer drainage to levels similar to thatin CC. After the midsummer drainage, DOC concentra-tion increased to 40 mg C L−1 at 72 DAT, with a con-comitant increase in CH4 flux (Figs. 1a and 2a).

In CC, NH4–N levels decreased from 38 mg N L−1 at7 DAT to near zero at 34 DAT (Fig. 2b). Concentrationsof NH4–N in RC and RI were less than 10mgN L−1 at 7DAT and reached near zero at 34 DAT. No obviouschanges were observed during the remaining period.NO3–N was not detected during the cultivation period(data not shown).

Figure 3a, b shows seasonal changes in the watertable and soil Eh at 5 cm soil depth, respectively. Beforethe midsummer drainage, soil Eh showed negativevalues; the average soil Eh was −109±21, −163±18,and −135±47 mV in CC, RC, and RI, respectively.During the midsummer drainage, soil Eh increased toapproximately +600 mV. After the midsummer drain-age, when continuous flooding in CC and RC andintermittent irrigation in RI began, the average soil Ehwas 31±102, −84±56, and 80±197 mV in CC, RC, andRI, respectively. In RC, soil Eh continued to decrease to−144 mV, while in RI, soil Eh increased from 72 DATand fluctuated around values higher than 0 mV.

Figure 4a shows correlation between DOC concen-tration and CH4 flux at different periods. In all sampledperiods, CH4 flux showed positive correlation withDOC concentration. In contrast, the slope of the regres-sion line became steeper from before-midsummer drain-age to after-midsummer drainage. Figure 4b shows therelationship between soil Eh and CH4 flux. Higher CH4

fluxes were distributed around −200 mV, especially in

Water Air Soil Pollut (2014) 225:2118 Page 5 of 13, 2118

RC and RI, whereas lower CH4 fluxes were found in soilwith higher Eh value.

3.4 mcrA Gene Copy Number

In all treatments, mcrA gene copy number increasedwith time (Fig. 5). However, these numbers differedamong treatments. At 34 DAT, there was no significantdifference between the treatments. In CC treatment, alarge increase in mcrA copy number was observed be-tween 79 and 113 DAT. After a rapid increase in CH4

flux in RC and RI on 72 DAT (Fig. 1), significantdifference in mcrA gene copy number was detected inCC and DTAD residue amended plots (RC and RI) on79 DAT.

3.5 Forage Rice Growth

Aboveground biomass sharply increased until harvest(Fig. 6a). On 34 DAT, the aboveground biomass was15.7, 3.3, and 2.3 g per plant in CC, RC, and RI,respectively. On 113 DAT, these values were 97.7,144, and 91.7 g per plant, respectively. Finally, the

average biomass of harvested plants was 76.5, 51.5,and 39.2 g per plant.

Root biomass showed a similar trend in all treatments(Fig. 6b). Maximum root weight was observed at 79DAT: 29.4, 23.1, and 15.6 g per plant in CC, RC, and RI,respectively. Afterwards, the root biomass decreased.

Table 2 shows the yield of rice determined by weightof 18 rice plants harvested from each plot. Treatment CChas the highest yield of 15.2 t·ha−1, followed by RC andRI. In RC and RI, the ratio of grain to total biomass wasalso lower than in CC.

4 Discussion

4.1 Influence of DTAD Residue Incorporationand Water Management on Dynamics of CH4 Flux

To date CH4 emissions from rice fields incorporatedwith wheat straw, rice straw, and composts have beenintensely studied (Ma et al. 2007; Wang et al. 2000;Watanabe et al. 1999; Yagi andMinami 1990). Values ofCH4 flux from forage rice fields incorporated withDTAD residue were consistent with those reported inprevious studies. We observed two active CH4 emissionstages; one is before midsummer drainage (7–43 DAT)and another is after midsummer drainage (45–97 DAT).Because organic matter is a substrate for CH4 produc-tion, DOC is an important indicator of CH4 emission(Lu et al. 2000b). However, in rice fields amended withorganic matter, the relationship between DOC concen-tration and CH4 flux has been rarely reported. In Fig. 4a,

Fig. 1 Seasonal changes in CH4 fluxes during the cultivation period.Double-headed arrows indicate drainage period. Intermittent irrigationwas carried out from 48 to 106 DAT in RI

Table 1 Cumulative greenhouse gas emissions during experi-mental period

Treatmenta CH4 emission (kg C ha−1) CO2-eqb (t CO2 ha

−1)

CC 105±38a 3.5±1.3a

RC 509±145b 17.0±4.8b

RI 306±44c 10.2±1.5c

2118, Page 6 of 13 Water Air Soil Pollut (2014) 225:2118

different slopes of the regression lines at different pe-riods suggest that the quality of DOC and environmentfor CH4 production changes with time. There have beenseveral reports of temporal changes in the source of CH4

emitted from rice fields incorporated with rice straw(Chidthaisong and Watanabe 1997; Watanabe et al.1998). Watanabe et al. (1999) found that the contribu-tion of incorporated rice straw to CH4 emission washigher in the early cultivation period, while in laterstages, the main source of CH4 shifted to plant-derivedmaterial such as root exudate and dead leaf or root.Before midsummer drainage (between 7–43 DAT), highDOC concentrations were detected in DTAD amendedplots (RC and RI; Fig. 2a) implying that the degradationof DTAD has occurred. In anaerobic soils, methanefermentation is a final process of a stepwise degradationof organic matter. During degradation, several kinds ofmonomers are produced by hydrolysis, acidogenesis,and acetogenesis (Le Mer and Roger 2001). Therefore,

a portion of DOC before midsummer drainage in RCand RI would not be directly related to CH4 production(Fig. 4a).

Low DOC concentration but rapid increase inCH4 flux after midsummer drainage suggest thatorganic matter in the pore water is converted moreefficiently into CH4 in DTAD residue-incorporatedplots. A concurrent increase in mcrA copy number,above-ground and root biomass, and CH4 flux in RCafter 72 DAT (Figs. 1, 5, and 6) suggests a linkbetween the population of methanogenic archaeaand plant growth and CH4 production. Rice rootsrelease exudates into the rhizosphere, and theamount of exudates increases with plant growth(Lu et al. 2000b). Methane production in anaerobicsoils amended with exudates was completed within5 days after treatment (Lu et al. 2000a). Thus, wespeculate that forage rice growth after midsummerdrainage in RC and RI provides exudates that are

Fig. 2 Seasonal changes in a DOC and b NH4–N concentrations at 5 cm soil depth. Double-headed arrows indicate drainage period.Intermittent irrigation was carried out from 48 to 106 DAT in RI

Water Air Soil Pollut (2014) 225:2118 Page 7 of 13, 2118

efficiently converted to CH4 in anaerobic soils rich inmethanogenic archaea. Furthermore, Watanabe et al.(1998) suggested that a part of the once assimilated

CO2–C, which originates from decomposition of incor-porated rice straw, would be emitted as CH4 in latergrowth stages. In addition, Yuan et al. (2014) reported

Fig. 3 Seasonal changes in a water table and b soil Eh at 5 cm soil depth. Double-headed arrows indicate drainage period. Intermittentirrigation was carried out from 48 to 106 DAT in RI

2118, Page 8 of 13 Water Air Soil Pollut (2014) 225:2118

Fig. 4 Relationship between CH4 flux and a DOC concentrationat different periods and b soil Eh in different treatments. MDindicates midsummer drainage. Dotted, gray, and black lines are

regression lines between DOC concentration and CH4 flux before,during, and after midsummer drainage, respectively

an increase in methanogenic archaea population afterrice straw addition, which enhanced CH4 production

from root organic carbon. Significantly higher mcrAcopy number in RC and RI than in CC at 79 DAT

Fig. 5 Seasonal changes in mcrA gene copy number. Different letters indicate significant differences among the treatments for eachsampling day (p<0.05)

Fig. 6 Changes in a above-ground biomass and b root biomass during cultivation period (n=3). Rice plants were harvested on 137 DAT.Different letters indicate significant differences between different sampling days within a treatment (p<0.05)

Water Air Soil Pollut (2014) 225:2118 Page 9 of 13, 2118

suggests that the incorporation of DTAD residue in-creases the population of methanogenic archaea similarto that by the addition of other organic material (ricestraw or compost application; Conrad and Klose 2006;Singh et al. 2012). Therefore, incorporation of DTADresidue plays an important role in CH4 emission byproviding organic C source for CH4 production as wellas in increase of methanogenic archaea population.

It has been well known that water management sig-nificantly affects the dynamics of CH4 flux in a ricefield. Methane flux substantially decreased after mid-summer drainage in all treatments, and it was lowerduring the intermittent irrigation in RI than in RC(Fig. 1), which is consistent with the previous studies(Ma et al. 2007; Tyagi et al. 2010; Yagi et al. 1996; Zouet al. 2005). In contrast to CH4 flux response to watermanagement, mcrA copy number continued to increasewith time in all treatments (Fig. 5). Further, the lack ofsignificant difference inmcrA copy number between RCand RI (Fig. 5) suggests that water management did notaffect the abundance of methanogenic archaea. Thisresult is in agreement with the dynamics of methano-genic archaea in intermittently irrigated rice fields(Watanabe et al. 2013).

Methanogenic activity might be more sensitive towater management than abundance. Midsummer drain-age and intermittent irrigation caused increase in soil Eh(Fig. 3b). As shown in Fig. 4b, lower CH4 flux wasobserved at higher soil Eh, which is consistent withprevious studies (Hou et al. 2000; Riya et al. 2012).From Fig. 2a, we can conclude that drainage inducesflow of atmospheric O2 into the soil, resulting in oxida-tion of organic matter and other material related to redoxstatus. In drained soil, where soil Eh values are higher,Fe3+ or SO4

2− are produced (Patrick and Jugsujinda

1992). In such soils, methanogenic archaea areoutcompeted by iron or sulfate reducers for organicmatter after flooding (Ma et al. 2012). Probably, soillayer above 5 cm was more likely to be important forCH4 control because soil Eh at soil–water interface ishigher than that at 5 cm depth during both flooded anddrained conditions (Hasebe et al. 1985). It has beenreported that after drainage vertical distribution of CH4

oxidation was extended from 2 to 8 mm, while CH4

production potential decreased in these layers (Henckelet al. 2001). Oxygen stress is also known to inhibitmethanogenic activity (Fetzer et al. 1993).

4.2 Influence of DTAD Residue Incorporationand Water Management on Dynamics of N2O Flux

To date, many researchers reported stimulation of N2Oemissions during drainage (Cai et al. 1999; Riya et al.2012; Zheng et al. 2000). In the present study, however,the observed N2O fluxes were below detection limitthroughout the cultivation period. Therefore, the posi-tive effect of intermittent irrigation on N2O emission isnegligible. Low N2O emissions were also confirmed byNishimura et al. (2004) and Yagi et al. (1996) in ricefields amended with ca. 100 kg N ha−1 of chemicalfertilizer. The observed lower NH4–N concentration inpore water (Fig. 2b) and significantly lower above-ground biomass in RC and RI compared to CC suggesta lack of available nitrogen in DTAD residue (Fig. 2b;Table 2). This may be because of a reduction in availablenitrogen as DTAD residue dried (see Subsection 2.2),leading to less N2O production in the soil during drain-age. Therefore, it is essential to improve the nutrientstatus of DTAD residue by composting and to evaluateN2O emission from a rice field treated with compostedDTAD residue.

4.3 Total Greenhouse Gas Emissions and its Mitigationby Water Management

In this study, the addition of DTAD residue significantlyincreased CH4 emission (Table 1). In rice farming, thepercentage of carbon emitted as CH4 from rice straw orcompost was 0.4–14 % or 2.1–3.1 %, respectively, ascalculated from data reported by Yagi and Minami(1990) and based on carbon content of 39 % and 31 %for rice straw and compost, respectively. In this study,carbon emitted as CH4 accounted for 9.3 % and 5.6 % ofresidues incorporated in RC and RI, respectively, similar

Table 2 Yield of harvested rice calculated from the weight of 18rice plants harvested from each plot

Treatmenta Yield (t·ha−1)b

CC 15.2±0.2 (44.2 %)a

RC 10.7±0.6 (36.3 %)b

RI 7.9±0.8 (28.2 %)c

Notes:a For explanation of treatments refer to Table 1bNumbers in parentheses indicate percentage of grain yield

Different lowercase letters indicate significant differences amongthe treatments (p<0.05)

2118, Page 10 of 13 Water Air Soil Pollut (2014) 225:2118

to the values reported for rice straw. Pre-treatment oforganic matter, such as composting, can reduce CH4

emissions compared with rice straw incorporation(Majumdar 2003; Pramanik and Kim 2014; Yagi andMinami 1990). Tsutsuki and Ponnamperuma (1987)reported 4.3–28 % and 0.3–8.7 % conversions fromtotal carbon to CH4 in rice straw and compost,respectively, under anaerobic soil conditions.Wassmann et al. (1993) reported that degradation ofreadily decomposable compounds by fermentation re-duced CH4 emissions from a rice field supplementedwith biogas residue. In this study, higher CH4 conver-sion of DTAD residue compared to compost, despite thedecomposition of 50 % of volatile solids during theDTAD process (see Subsection 2.2), suggests thatDTAD residue of pig manure and forage rice straw usedin the present study had a relatively higher content ofdecomposable organic matter compared to compost.

Water management can effectively mitigate CH4

emissions. In rice paddy fields, 42–74 % of CH4 emis-sions were mitigated by drainage (Minamikawa andSakai 2006; Yagi et al. 1996; Zou et al. 2005). Compar-ison of CH4 emissions between RC and RI showed thatintermittent irrigation reduced CH4 emissions by 40 %,which is consistent with the results reported in previousstudies. However, the cumulative CH4 emission in RIwas still three times higher than in CC, indicating thataddition of DTAD residue to forage rice fields canincrease greenhouse gas emissions compared to theconventional rice management. Further research shouldfocus on analyzing greenhouse gas emissions through-out the entire pig manure management process, includ-ing pig farm, DTAD process, and application to foragerice fields, and compare it against the conventional pigmanure management system.

5 Conclusions

This study is the first report of CH4 and N2O emissionsfrom forage rice field amended with DTAD residue andits mitigation by water management. Addition of DTADresidue strongly increased CH4 emissions throughoutthe entire cultivation period by providing carbon sourcefor CH4 production and increase in methanogenic ar-chaea population. Intermittent irrigation reduced CH4

emissions by 40 % without increasing N2O emissions.Thus, intermittent irrigation is an effective approach for

mitigating CH4 emission in forage rice fields supple-mented with DTAD residue.

Acknowledgments This study was supported by an Environ-ment Research & Technology Development Fund (B-1103) fromtheMinistry of the Environment, Japan.We thank two anonymousreferees for their valuable comments and constructive suggestions.We are also grateful to Mr. Tamotsu Ishibashi, Kurita WaterIndustries LTD., for providing inoculum for dry-thermophilicanaerobic digestion.

References

Arthurson, V. (2009). Closing the global energy and nutrientcycles through application of biogas residue to agriculturalland—potential benefits and drawbacks. Energies, 2(2), 226–242.

Cai, Z. C., Xing, G. X., Shen, G. Y., Xu, H., Yan, X. Y., Tsuruta,H., et al. (1999). Measurements of CH4 and N2O emissionsfrom rice paddies in Fengqiu, China. Soil Science and PlantNutrition, 45(1), 1–13.

Chidthaisong, A., & Watanabe, I. (1997). Methane formationand emission from flooded rice soil incorporated with13C-labeled rice straw. Soil Biology and Biochemistry,29(8), 1173–1181.

Conrad, R., & Klose, M. (2006). Dynamics of the methanogenicarchaeal community in anoxic rice soil upon addition ofstraw. European Journal of Soil Science, 57(4), 476–484.

Fetzer, S., Bak, F., & Conrad, R. (1993). Sensitivity of methano-genic bacteria from paddy soil to oxygen and desiccation.FEMS Microbiology Ecology, 12(2), 107–115.

Hasebe, A., Sekiya, S., & Iimura, K. (1985). Direct determinationof the differentiation process of the oxidized and reduced soillayers in paddy fields. Jarq-Japan Agricultural ResearchQuarterly, 19(3), 172–177.

Henckel, T., Jackel, U., & Conrad, R. (2001). Vertical distributionof the methanotrophic community after drainage of rice fieldsoil. FEMS Microbiology Ecology, 34(3), 279–291.

Hou, A. X., Chen, G. X., Wang, Z. P., Van Cleemput, O., &Patrick, W. H. (2000). Methane and nitrous oxide emissionsfrom a rice field in relation to soil redox and microbiologicalprocesses. Soil Science Society of America Journal, 64(6),2180–2186.

IPCC. (2007). Climate Change 2007: The Physical Science Basis.In S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis,K. B. Averyt, et al. (Eds.), Contribution of Working Group Ito the Fourth Assessment Report of the IntergovernmentalPanel on Climate Change (p. 996). United Kingdom andNew York: Cambridge University Press.

Jha, A. K., Li, J., Nies, L., & Zhang, L. (2011). Research advancesin dry anaerobic digestion process of solid organic wastes.African Journal of Biotechnology, 10(65), 14242–14253.

Jiao, Z. H., Hou, A. X., Shi, Y., Huang, G. H., Wang, Y. H., &Chen, X. (2006). Water management influencing methaneand nitrous oxide emissions from rice field in relation to soilredox and microbial community. Communications in SoilScience and Plant Analysis, 37(13–14), 1889–1903.

Water Air Soil Pollut (2014) 225:2118 Page 11 of 13, 2118

Kyaw, K. M., Toyota, K., Okazaki, M., Motobayashi, T., &Tanaka, H. (2005). Nitrogen balance in a paddy field plantedwith whole crop rice (Oryza sativa cv. Kusahonami) duringtwo rice-growing seasons. Biology and Fertility of Soils,42(1), 72–82.

Le Mer, J., & Roger, P. (2001). Production, oxidation, emissionand consumption of methane by soils: a review. EuropeanJournal of Soil Biology, 37(1), 25–50.

Lu, Y., Wassmann, R., Neue, H. U., Huang, C., & Bueno, C. S.(2000a). Methanogenic responses to exogenous substrates inanaerobic rice soils. Soil Biology and Biochemistry, 32(11–12), 1683–1690.

Lu, Y. H., Wassmann, R., Neue, H. U., & Huang, C. (2000b).Dynamics of dissolved organic carbon and methane emis-sions in a flooded rice soil. Soil Science Society of AmericaJournal, 64(6), 2011–2017.

Ma, J., Li, X. L., Xu, H., Han, Y., Cai, Z. C., & Yagi, K. (2007).Effects of nitrogen fertiliser and wheat straw application onCH4 and N2O emissions from a paddy rice field. AustralianJournal of Soil Research, 45(5), 359–367.

Ma, K., Conrad, R., & Lu, Y. (2012). Responses of methanogenmcrA genes and their transcripts to an alternate dry/wet cycleof paddy field soil. Applied and EnvironmentalMicrobiology, 78(2), 445–454.

Majumdar, D. (2003). Methane and nitrous oxide emission fromirrigated rice fields: proposed mitigation strategies. CurrentScience, 84(10), 1317–1326.

Minamikawa, K., & Sakai, N. (2006). The practical use of watermanagement based on soil redox potential for decreasingmethane emission from a paddy field in Japan. Agriculture,Ecosystems and Environment, 116(3–4), 181–188.

Nishimura, S., Sawamoto, T., Akiyama, H., Sudo, S., & Yagi, K.(2004).Methane and nitrous oxide emissions from a paddy fieldwith Japanese conventional water management and fertilizerapplication. Global Biogeochemical Cycles, 18(2), GB2017.

Patrick, W. H., & Jugsujinda, A. (1992). Sequential reduction andoxidation of inorganic nitrogen, manganese, and iron inflooded soil. Soil Science Society of America Journal,56(4), 1071–1073.

Pramanik, P., & Kim, P. (2014). Evaluating changes in cellulolyticbacterial population to explain methane emissions from air-dried and composted manure treated rice paddy soils. Scienceof the Total Environment, 470(471), 1307–1312.

Riya, S., Zhou, S., Watanabe, Y., Sagehashi, M., Terada, A., &Hosomi, M. (2012). CH4 and N2O emissions from differentvarieties of forage rice (Oryza sativa L.) treating liquid cattlewaste. Science of the Total Environment, 419, 178–186.

Sanchis, E., Ferrer, M., Torres, A. G., Cambra-López, M., &Calvet, S. (2012). Effect of water and straw managementpractices on methane emissions from rice fields: a reviewthrough a meta-analysis. Environmental EngineeringScience, 29(12), 1053–1062.

Sasada, Y., Win, K. T., Nonaka, R., Win, A. T., Toyota, K.,Motobayashi, T., et al. (2011). Methane and N2O emis-sions, nitrate concentrations of drainage water, and zincand copper uptake by rice fertilized with anaerobicallydigested cattle or pig slurry. Biology and Fertility ofSoils, 47(8), 949–956.

Singh, A., Singh, R. S., Upadhyay, S. N., Joshi, C. G., Tripathi, A.K., & Dubey, S. K. (2012). Community structure of methan-ogenic archaea and methane production associated with

compost-treated tropical rice-field soil. FEMS MicrobiologyEcology, 82(1), 118–134.

Steinberg, L. M., & Regan, J. M. (2009). mcrA-targeted real-timequantitative PCR method to examine methanogen communi-ties. Applied and Environmental Microbiology, 75(13),4435–4442.

Tsutsuki, K., & Ponnamperuma, F. N. (1987). Behavior ofanaerobic decomposition products in submerged soils:effects of organic material amendment, soil properties,and temperature. Soil Science and Plant Nutrition,33(1), 13–33.

Tyagi, L., Kumari, B., & Singh, S. N. (2010). Water manage-ment—a tool for methane mitigation from irrigated paddyfields. Science of the Total Environment, 408(5), 1085–1090.

US-EPA (2012) Global Anthropogenic Non-CO2 Greenhouse GasEmissions: 1990–2030. Washington, U.S. http://www.epa.gov/climatechange/Downloads/EPAactivities/EPA_Global_NonCO2_Projections_Dec2012.pdf

Wang, Z. Y., Xu, Y. C., Li, Z., Guo, Y. X., Wassmann, R.,Neue, H. U., et al. (2000). A four-year record of methaneemissions from irrigated rice fields in the Beijing regionof China. Nutrient Cycling in Agroecosystems, 58(1–3),55–63.

Wassmann, R., Wang, M. X., Shangguan, X. J., Xie, X. L., Shen,R. X., Wang, Y. S., et al. (1993). First records of a fieldexperiment on fertilizer effects on methane emission fromrice fields in Hunan-province (PR China). GeophysicalResearch Letters, 20(19), 2071–2074.

Watanabe, A., Yoshida, M., & Kimura, M. (1998). Contribution ofrice straw carbon to CH4 emission from rice paddies using13C-enriched rice straw. Journal of Geophysical Research,[Atmospheres], 103(D7), 8237–8242.

Watanabe, A., Takeda, T., & Kimura, M. (1999). Evaluation oforigins of CH4 carbon emitted from rice paddies. Journal ofGeophysical Research, [Atmospheres], 104(D19), 23623–23629.

Watanabe, T., Hosen, Y., Agbisit, R., Llorca, L., Katayanagi, N.,Asakawa, S., et al. (2013). Changes in community structureof methanogenic archaea brought about by water-savingpractice in paddy field soil. Soil Biology and Biochemistry,58, 235–243.

Win, K. T., Nonaka, R., Toyota, K., Motobayashi, T., & Hosomi,M. (2010). Effects of option mitigating ammonia volatiliza-tion on CH4 and N2O emissions from a paddy field fertilizedwith anaerobically digested cattle slurry. Biology andFertility of Soils, 46(6), 589–595.

Yagi, K., & Minami, K. (1990). Effect of organic matter applica-tion on methane emission from some Japanese paddy fields.Soil Science and Plant Nutrition, 36(4), 599–610.

Yagi, K., Tsuruta, H., Kanda, K., & Minami, K. (1996). Effect ofwater management onmethane emission from a Japanese ricepaddy field: Automated methane monitoring. GlobalBiogeochemical Cycles, 10(2), 255–267.

Yuan, Q., Pump, J., & Conrad, R. (2014). Straw application inpaddy soil enhances methane production also from othercarbon sources. Biogeosciences, 11, 237–246.

Zheng, X., Wang, M., Wang, Y., Shen, R., Gou, J., Li, J., et al.(2000). Impacts of soil moisture on nitrous oxide emissionfrom croplands: a case study on rice-based agro-ecosystem inSoutheast China. Chemosphere Global Change Science,2(2), 207–224.

2118, Page 12 of 13 Water Air Soil Pollut (2014) 225:2118

Zhou, S., Kanai, R., Suzuki, K., Riya, S., Terada, A., & Hosomi,M. (2013). Dry-thermophilic anaerobic co-digestion of swinemanure mixed with forage rice straw. In InternationalConference on Recent Advances in Pollution Control andResource Recovery for the Livestock Farming Industry(RAPCP 2013), Jiaxing, China, (pp. 247–250).

Zou, J., Huang, Y., Jiang, J., Zheng, X., & Sass, R. L.(2005). A 3-year field measurement of methane andnitrous oxide emissions from rice paddies in China:Effects of water regime, crop residue, and fertilizerapplication. Global Biogeochemical Cycles, 19(2),GB2021.

Water Air Soil Pollut (2014) 225:2118 Page 13 of 13, 2118