Measurement of nitrous oxide emissions from two rice ... - Springer Link

Nutrient Cycling in Agroecosystems 64: 125–133, 2002. © 2002 Kluwer Academic Publishers. Printed in the Netherlands.

125

Measurement of nitrous oxide emissions from two rice-based cropping systems in China Z.Q. Xiong1,∗ , G.X. Xing1 , H. Tsuruta2 , G.Y. Shen1 , S.L. Shi1 & L.J. Du1 1 Institute

of Soil Science, Academia Sinica, Nanjing 210008, P.R. China; Environmental Sciences, Tsukuba 305, Japan (∗ Corresponding author)

2 National

Institute of Agro-

Key words: cropping system, fallow, fertilization, green manure, N2 O emission, paddy

Abstract A field experiment was conducted to investigate the effects of winter management and N fertilization on N2 O emission from a double rice-based cropping system. A rice field was either cropped with milk vetch (plot V) or left fallow (plot F) during the winter between rice crops. The milk vetch was incorporated in situ when the plot was prepared for rice transplanting. Then the plots V and F were divided into two sub-plots, which were then fertilized with 276 kg urea-N ha−1 (referred to as plot VN and plot FN) or not fertilized (referred to as plot VU and plot FU). N2 O emission was measured periodically during the winter season and double rice growing seasons. The average N2 O flux was 11.0 and 18.1 µg N m−2 h−1 for plot V and plot F, respectively, during winter season. During the early rice growing period, N2 O emission from plot VN averaged 167 µg N m−2 h−1 , which was eightto fifteen-fold higher than that from the other three treatments (17.8, 21.0 and 10.8 µg N m−2 h−1 for plots VU, FN, and FU, respectively). During the late rice growing period, the mean N2 O flux was 14.5, 11.1, 12.1 and 9.9 µg N m−2 h−1 for plots VN, VU, FN and FU, respectively. The annual N2 O emission rates from green manure-double rice and fallow-double rice cropping systems were 3.6 kg N ha−1 and 1.3 kg N ha−1 , respectively, with synthetic N fertilizer, and were 0.99 kg N ha−1 and 1.12 kg N ha−1 , respectively, without synthetic N fertilizer. Generally, both green manure N and synthetic fertilizer N contribute to N2 O emission during double rice season. Introduction Nitrous oxide, a greenhouse gas in the atmosphere, has a global averaged atmospheric concentration of about 317 ppbv (CMDL, 2000). It is also an important source of stratospheric nitrogen oxides that start a chain of reactions leading to stratospheric ozone destruction (Crutzen, 1991). Measurements of air occluded in polar ice cores show that N2 O has been increasing for at least the past 100 years (Khalil and Rasmussen, 1988). Worldwide measurements have revealed that the concentration is increasing at an annual rate of 0.2–0.3% (Weiss, 1981; Prinn et al., 1990). The recent increase is considered to be due to the increase in anthropogenic N sources such as synthetic nitrogen fertilizer, increased biological N-fixation and fossil fuel combustion (Mosier and Kroeze, 2000). Anthropogenic sources of N2 O are largely biogenic with agriculture as a major contributor (Mosier et al., 1996). Biogenic production of N2 O in the soil res-

ults primarily from the nitrification and denitrification processes and major factors governing these processes are carbon and nitrogen substrate availability, soil moisture and temperature (Mosier et al., 1998). Soil management practices, through their effect on these factors, can indirectly influence N2 O formation and release from soils (Jacinthe and Dick, 1997). Reports on effects of legumes on N2 O emission from agricultural soils and pastures have shown that legumes may contribute to N2 O fluxes directly or indirectly (Bouwman and Sombroek, 1990; Galbally et al., 1992). Green legume crop residues, at early stages of growth, decompose rapidly after incorporation into soil and release not only mineral N but also supply considerable organic C to soil microorganisms. Thus, the addition of legume residues could increase the loss of N through denitrification if soil is relatively wet or anaerobic (John et al., 1989; Andren et al., 1990; Aulakh et al., 1991). Several studies have reported on the influence of different crop residues on biochemical

126

Figure 1. Temporal variations of N2 O emissions (a), soil temperature and soil water content (b) during winter season. Soil water content is the mean of measurements within each treatment (n = 3) and soil temperature is the mean of temperatures measured in all plots (n = 6); Bars indicate standard deviation; Treatments are defined in Table 1. Table 1. Description of experimental treatments for measurements of N2 O emissions in rice-based cropping systems Treatment

Crop season Winter season Description

F

Fallow, no K or P fertilizers

V

Milk vetch, 124 kg K2 O·ha−1 and 240 kg P2 O5 ·ha−1

Late rice

Treatment

Early ricea Description

FU FN VU VN

No N fertilizer Applied at 276 kg N ha−1 Vetch at 124 kg N ha−1 , no N fertilizer Vetch at 124 kg N ha−1 , Applied at 276 kg N ha−1

No N fertilizer Applied at 276 kg N ha−1 No N fertilizer Applied at 276 kg N ha−1

a All treatments during the rice growing periods received 90 kg K O·ha−1 and 192 kg P O ·ha−1 as basal fertilizer. 2 2 5

N transformations such as mineralization, immobilization, and denitrification (Beauchamp et al., 1989; Aulakh et al., 1983a,b, 1984a, 1991; McKenney et al., 1993). Biologically fixed nitrogen by legume crops is considered as a source of N2 O emission by IPCC (1996). However, information on effect of leguminous green manure in conjunction with synthetic N fertilizer on N2 O emission is very limited (Aulakh et al., 2001). China is one of the major rice cultivation countries in the world with an area of 2.5×107 ha, accounting for 23% of the world’s total paddy field area. The percentage of double rice cropping region amounted to 66% of the total rice paddy area in China (Li, 1992). In this region, there are two typical cropping systems: double rice followed by fallow in winter and double rice followed by milk vetch (Astragalus sinicus L.) in winter. A field experiment was conducted in a subtropical region of Jiangxi, China. With winter fallow-double rice system as control, the effects of: (1) milk vetch cultivation in winter, (2) milk vetch incorporated as green manure, (3) N fertilization, and (4) green ma-

nure N combined with synthetic fertilizer nitrogen on nitrous oxide emissions during the following rice growing period were investigated in this paper.

Materials and methods Experimental site Field experiments were conducted in a paddy field at the Red Soil Ecological Experiment Station in Yingtan, Jiangxi province, China (28◦ 15 N, 116◦55 E). This region is a typical sub-tropic climate zone. A mean annual rainfall of 1795 mm is unevenly distributed throughout the year, of which 50% occurs from April to June. Annual mean air temperature is 17.6 ◦ C. The experimental soil was classified as Hydragric Anthrosols developed from Quaternary red clay, and taken at 0–15 cm depth before the experiment. The soil sample contained 0.76 g kg−1 total N, and 14.0 g kg−1 organic C, and had a pH (H2 O) of 5.52. Before conducting the experiment, the field

127

Figure 2. Temporal variations of N2 O emissions from the plot VN (a) and plots VU, FU, and FN (b), soil redox potential (c), and soil temperature (d) during the early rice growing season. Treatments are defined in Table 1; BF, basal fertilizer; TP, transplanting; TD, top-dressing; MSA, mid-season aeration. Soil redox potential values are the mean of values measured within each treatment (n = 3) and soil temperature is the mean of all temperature measurements within all plots (n = 12). Bars indicate standard error for (a) and (b) and standard deviation for (c) and (d).

was under an intensive cropping rotation of winter fallow-rice-rice cropping system for at least 10 years. Experimental treatments After the late rice harvest in 1999, a field was either fallowed (plot F) or planted to milk vetch (plot V). The size of each main plot was 8 m×7 m. Drainage was performed for the previous late rice harvest. Twentyfive days prior to early rice transplanting, fresh milk vetch was incorporated into soil in situ in plot V at a rate of 33 750 kg ha−1 fresh weight, containing 124 kg N ha−1 (C:N=16.5). Then plot V as well as plot F was divided into two sub-plots (4 m×7 m for each sub-plot). Subplots received urea fertilizer at rates of 0 (VU and FU) or 276 kg N ha−1 (VN and FN) per rice season. Tripplicate plots were randomly assigned within the four treatments during both early rice and late rice growing season (see Table 1). Urea was broadcast as basal fertilizer on transplanting day and top dressing was applied 8–9 days later evenly

in N fertilization treatments (plots VN and FN). Rice seedlings of 45-day-old were transplanted at 20×20 cm spacing. Adopted from the local farmers’ practices, a standing water layer about 3–5 cm in depth was maintained after rice transplanting till mid-season aeration (MSA). Then, alternative drying and flooding and final drainage was practiced. N2 O flux, soil temperature at 5-cm depth, and soil redox potential (Eh) at 10-cm depth (or soil water content in % (w/w)) were measured simultaneously. N2 O flux measurement N2 O flux was measured with closed chambers (Terry et al., 1981). N2 O flux measurements started on the day of sowing milk vetch seeds and ended on the day of milk vetch incorporation in the winter season (Figure 1) and started 2 days after rice transplanting and ended on the harvest during the rice growing period (Figures 2 and 3). Fluxes of N2 O from all treatments were measured generally once a week ex-

128 Table 2. N2 O emission rate during each cropping season as affected by crop season and fertilization (1999–2000, Yingtan, Jiangxi) Crop Seasona

Treatmentb

N2 O flux (µg N m−2 h−1 ) Average±SD Max. Min.

N2 O-N (kg ha−1 )

% of N added

Winter season (163)

V F

11.0±2.4 18.1±3.6

27.2 65.5

2.72 9.26

0.43±0.09 bc 0.71±0.14 a

/ /

Early rice (70)

VN VU FN FU

167±98.6 17.8±3.6 21.0±4.3 10.8±2.4

828 80.7 70.8 23.5

7.20 −5.43 2.63 2.67

2.81±1.66 a 0.30±0.06 b 0.35±0.07 b 0.18±0.04 b

0.66 0.094 0.062 /

Late rice (98)

VN VU FN FU

14.5±3.8 11.1±2.6 12.1±2.2 9.9±2.2

46.0 34.2 34.2 30.3

1.74 3.10 3.83 1.26

0.34±0.09 a 0.26±0.06 bc 0.28±0.05 b 0.23±0.05 c

0.040 / 0.019 /

a Values in parentheses were length of growing period. b Treatment symbols were defined in Table 1. c For each crop season, means±standard deviation (SD) followed by the same letters did not

differ significantly within the same column at P < 0.05 level according to Duncan’s new multiple range test.

cept in December, January and February when fluxes were measured at 15-day interval due to cold weather. Measurements were made more frequently after basal fertilization and top-dressing and during mid-season aeration (MSA). Gas samples were collected using 50×50 cm rectangular chambers made from plexiglass. Five hills of rice seedlings were covered in each sampling chamber. For any one N2 O flux measurement, four gas samples from each chamber were withdrawn at 10-min interval. The gas samples were analyzed for N2 O concentration by gas chromatography (HP 5890 II) equipped with electron capture detector (ECD). Cumulative N2 O emissions were calculated from the individual fluxes and the time between the measurements. Averages of N2 O flux were calculated from the individual fluxes weighed by the time between the measurements. Results were given as means with standard deviation and difference between treatments within each crop season was analyzed according to Duncan’s new multiple range test at P < 0.05.

Results N2 O flux during the winter season Nitrous oxide fluxes were greatest immediately after initiating the study (Figure 1). The fallow plot (plot F)

exhibited greater fluxes and temporal variations than the milk vetch plot (Plot V). Both plots did not show any lag in N2 O emission following the beginning of this winter sampling period and emissions decreased gradually to a very low rate after about one month of sampling time (Figure 1a). Soil temperature decreased from approximately 18 ◦ C to 13 ◦ C during that time. N2 O emission from plot V had an average CV of 27.2% across the sampling period while the CV of N2 O emissions from plot F was 19.7%. That N2 O fluxes from plot F were generally higher than from plot V during the winter observation period (Figure 1a) was consistent with the difference of soil water contents between the two plots (Figure 1b). However, there was no significant correlation observed between N2 O emission and neither soil water content in plot V and plot Fnor soil temperature in plot V (Table 3). Significantly positive correlation was found only between N2 O emission and soil temperature in plot F (Table 3). Over the 163-day winter season, averaged N2 O flux was 11±2.4 and 18±3.6 µg N m−2 h−1 , with cumulative emission rate of 0.43±0.09 and 0.71±0.14 kg N ha−1 from plot V and plot F, respectively. N2 O emissions from plot V were 39% less (Table 2) than from plot F (P < 0.05).

129

Figure 3. Temporal variations of N2 O emissions (a), soil redox potential (b), and soil temperature (c) during the late rice growing season. Treatments are defined in Table 1; BF, basal fertilizer; TP, transplanting; TD, top-dressing; MSA, mid-season aeration; Soil redox potential is the mean of measurements within each treatment (n = 3) and soil temperature is the mean of temperature measurements from all plots (n = 12); Bars indicate standard deviation.

N2 O emissions during the early rice growing period During the following rice growing period, nitrous oxide fluxes tended to occur largely in separated peak periods (Figure 2). Among the four treatments, plot VN, which integrated milk vetch green manure with synthetic N fertilizer, emitted eight- to fifteen-fold more N2 O than the other three treatments (Figure 2a and 2b and Table 2). The fluxes from plot VN were especially high during and after the mid-season aeration period. The mean flux from plot VN was 167±98.6 µg N m−2 h−1 , accounting for an average of 0.66% of the added nitrogen during the early rice growing period. In the other three treatments, there appeared a peak emission after the basal fertilization and remained quite steady emission during the other observation period (Figure 2a and 2b). The mean N2 O fluxes from plots FN, FU, and VU were 21±4.3, 11±2.4 and 18±3.6 µg N m−2 h−1 , respectively. Fertilizer-induced N2 O-N emission from plots VU and FN treatment was 0.094% and 0.062%, respectively (Table 2). As the N application rate was not the same for green manure and

urea fertilizer treatments, the comparison of fertilizer induced N2 O-N emission is not valid. The temporal variation trend of N2 O flux from plot VN was similar to that of soil Eh (Figure 2a and 2c). After the mid-season aeration, the Eh increased dramatically and remained positive. N2 O emission from plot VN was significantly correlated to soil redox potential and soil temperature (P < 0.01). The emission from plot FU was correlated to soil temperature (Table 3). The spatial variability of N2 O fluxes was low in all plots (20–22% CV) except plot VN, which had an average CV of 59% (see Table 2). N2 O emissions during the late rice growing period During the late rice growing season, temporal variation patterns and magnitude of N2 O fluxes were similar among the four treatments. Peak fluxes were observed immediately after basal and top-dressing fertilizer additions, and then fluxes decreased to a very low level until MSA. After MSA, the fluxes remained relatively high till harvesting (Figure 3). The N2 O

130 Table 3. Coefficients of correlation between N2 O flux and soil temperature, redox potential, or water content during the winter, early rice, and late rice growing period Crop Season

Treatmenta

Temperature

Eh

Water content

Winter season

V F

0.1692ns 0.5181∗∗

ND ND

−0.2891ns 0.2008ns

Early rice

VN VU FN FU

0.3215∗ 0.0948ns 0.1624ns 0.4173∗

0.7673∗∗ −0.0707ns −0.0049ns 0.1642ns

ND ND ND ND

Late rice

VN VU FN FU

0.0110ns 0.0847ns −0.1592ns −0.2651ns

−0.1146ns −0.1212ns −0.0462ns 0.3793∗

ND ND ND ND

a Treatment symbols were defined in Table 1. ∗ , ∗∗ , ns mean differed significantly at P < 0.05, P < 0.01 and

P > 0.05, respectively. ND, Not determined.

fluxes were neither significantly correlated with soil temperature nor soil redox potential in all the plots except between N2 O flux and Eh in plot FU (P < 0.05, Table 3). The seasonal N2 O emission from plot VN during the late rice growing period (0.34±0.09 kg N ha−1 ) was much less than that during the early rice growing period (2.81±1.66 kg N ha−1 ), although it was still significantly higher than those from other treatments (P < 0.05, Table 2). A significant difference of seasonal N2 O emission was also observed between plot FN and plot FU (P < 0.05, Table 2). Compared to the treatments without N application (plot VU and plot FU), the corresponding percentage of seasonal N2 O emission to the added N fertilizer was still higher in plot VN (0.04%) than in plot FN (0.02%, Table 2). The result suggested that a small residual effect of milk vetch and N fertilization from the early rice growing period continued through the late rice cropping period. Annual N2 O emission Milk vetch and fallow in winter followed by double rice with N fertilization were two conventional ricebased cropping systems in the experimental site. Two other treatments without N fertilizer were adopted to compare the effects of green manure and N fertilization. Annual emission was calculated from Table 2 as the sum of the emission from the three crop-

growing seasons. The annual N2 O emission from green manure-double rice and fallow-double rice cropping systems were 3.6±1.8 kg N ha−1 and 1.3±0.26 kg N ha−1 , respectively, with synthetic N fertilizer, and 0.99±0.21 kg N ha−1 and 1.12±0.23 kg N ha−1 , respectively, without synthetic N fertilizer. Milk vetch, as green manure, significantly increased N2 O emission especially when used in conjunction with synthetic N fertilization. However, if green manure is incorporated into soils, farmers usually reduce the amount of N fertilizer so that the total N application from chemical fertilizer and green manure is comparable to the amount of chemical fertilizer alone. From plots for the FU, VU, FN and VN plots, rice total yields were 4500, 6000, 9250 and 10 400 kg ha−1 , respectively, in the double rice season, and the yields were lower in the early rice season than in late rice season.

Discussion Effect of milk vetch on N2 O emission in winter season Both plot V and plot F emitted higher N2 O during the first 16 days of the winter sampling period (Figure 1). Change from flooded rice to winter season enhanced N2 O release after the soil was drained. Relatively high temperature during this period promoted N2 O production. A residual effect of fertilization in the previous late rice season probably existed, and drainage and cultivation intensify the N turnover in soils (Bøckman et al., 1998). All of the mentioned above could contribute to the relatively high N2 O emission during the first sampling phase. Many reports (e.g. Bremner et al., 1980; Duxbury et al., 1982; Galbally et al., 1992; Carran et al., 1995; Plant and Bouman, 1999) indicate that leguminous crops increase N2 O emissions from soils because the plants biologically fix atmospheric nitrogen into soils. But the comparisons to non-legume crops or to uncultivated soil were not included in those experiments. While, Jacinthe and Dick (1997) found that without N fertilizer application, seasonal N2 O-N losses were lower in ridge-till soybean plot than that in no-till wheat plot. They did not provide the reasons for N2 O reduction in ridge-till soybean plot. MacKenzie et al. (1997, 1998) found that a corn system using legumes in rotation and moderate fertilizer would reduce N2 O emission but without comparison non-legume of corn at the same N level. These rotation systems had just

131 one cropping in a year and covered several years for a rotation. Inclusion of legumes into cropping systems should reduce fertilizer N requirements compared with continuous corn and N2 O emission under legume-corn rotations may be lower than under continuous corn (Jacinthe and Dick, 1997; MacKenzie et al., 1997, 1998). Seasonal emissions of N2 O from the milk vetch plot (0.43 kg N2 O-N ha−1 ) obtained from this experiment was comparable to the annual emissions (0.34–1.97 kg N2 O-N ha−1 ) reported by Bremner et al. (1980). Our results showed that in winter in the subtropical region, milk vetch cultivation significantly reduced N2 O emission by 39% compared to winter fallow. Residual N from the previous late rice season may result in the initial vigorous growth of milk vetch and then milk vetch utilizes N mainly from soil N. Lower soil moisture probably played a role in lower N2 O emissions from the milk vetch plots. No-till fallow soils had greater water content than milk vetch cultivation soil (Figure 1b). This is similar to the result of Aulakh et al. (1984b) who found greater soil water content in zero-till than in conventional-tillage with wheat. Effects of leguminous green manure on N2 O emission in rice growing season Incorporation of plant residues has frequently been observed to promote N loss (John et al., 1989; Andren et al., 1990; Aulakh et al., 1991; McKenney et al., 1993; Ragab et al., 1994; Allen et al., 1996; Lessard et al., 1996; Cochran et al., 1997; Smith et al., 1998). In contrast, however, Baggs et al. (2000) reported that N2 O evolution was not greatly influenced by the incorporated green manures. Cai et al. (2001) found that straw addition didn’t affect N2 O evolution in rice fields, either under continuous flooding or flooded/drained cycles. In the study reported here, seasonal N2 O emission from the plot incorporated with milk vetch (plot VU) during the early rice growing period was 0.30±0.06 kg ha−1 , 67% higher than that from the plot FU without incorporation of milk vetch. The difference was, however, not statistically significant (P > 0.05) because of the large variation of replicate measurements (Table 2). The difference of the N2 O emissions from the VU and FU plots became smaller during the following late rice growing period, suggesting that the residual effect of milk vetch on N2 O emissions was negligible during the second rice growing season when no additional N fertilizer

Figure 4. Distribution of total N2 O emissions in the different rice growing stages during the double rice growing period. Treatments are defined in Table 1. BF, from basal fertilization till top-dressing; TD, from top-dressing till mid-season aeration; MSA, from mid-season aeration till harvesting.

was applied. However, when N fertilizer was applied at 276 kg N ha−1 , seasonal N2 O emission from the plot incorporated with milk vetch (plot VN) during the early rice growing period was 2.81 kg N ha−1 , eight-fold higher than that from the plot FN without incorporation of milk vetch (0.35 kg N ha−1 ) (Table 2). The difference of the N2 O emissions from the VN and FN plots was still significant (P < 0.05) during the following late rice growing period though smaller than that during the early rice period. There was a trend that seasonal N2 O emission during the rice growing period increased with increasing N application rate (Table 2). The result suggests that milk vetch green manure significantly increased N2 O emission when in conjunction with N fertilization during both rice seasons. Effect of water regime on seasonal distribution of N2 O emissions N2 O emissions were greatly influenced by fertilizer application and water regime (Figures 2 and 3). Three stages of peak N2 O emissions are defined in Figure 4 according to agricultural management practices, namely, BF, TD and MSA stage. BF began with transplanting and basal fertilization until top-dressing; TD began with top-dressing until MSA drainage. MSA started at mid-season drainage till harvest. Figure 4 shows that N2 O emissions were different among growing stages. All treatments in early and late rice, except for VN in early rice season which emitted 95% after MSA, emitted only about a half of N2 O after midseason aeration. About one half of N2 O was emitted

132 during the continuous flooding period, which is consistent with the relative length of this period to the whole sampling period. Fourteen–26% of the total N2 O emission was emitted during the period from the basal fertilization to top-dressing.

Summary N2 O emission was dramatically increased by the application of milk vetch and synthetic N fertilizer together during the early rice growing period. The ratio of N2 O emission to applied N in plot VN was obviously higher than that in the other plots (Table 2). The emission pattern was also changed by the mixed application of milk vetch and synthetic fertilizer (Figure 4). About 92% of the N2 O emission took place after the MSA drainage till harvesting in plot VN, while only about half emissions took place in the other treatments during the same period. During the early rice growing period, the total N application rate was approximately 400 kg N ha−1 . Velthof et al. (1996) observed that the ratio of N2 O emission to applied N was higher in the treatment with high N application rate than in the treatment with low rate. Therefore, the increase in the ratio of N2 O emission to applied N in plot VN during the early rice growing period could be partly attributed to the high N application rate.

Acknowledgements This research was financially supported by the National Natural Science Foundation of China (Grant No. 39790100). We thank Dr. Z.C. Cai for his kind help in writing of the manuscript.

References Allen AG, Jarvis SC & Headon DM (1996) Nitrous oxide emissions from soils due to inputs of nitrogen from excreta return by livestock on grazed grassland in the UK. Soil Biol Biochem 28: 597–607 Andren O, Lindberg T, Paustian K & Rosswall T (1990) Ecology of arable land: organisms, carbon and nitrogen cycling. Ecol Bull 40: 127–152 Aulakh MS, Doran JW, Waiters DT, Mosier AR & Francis DD (1991) Crop residue type and placement effects on denitrification and mineralization. Soil Sci Soc Am J 55: 1020–1025 Aulakh MS, Khera TS & Doran JW (2001) Denitrification, N2 O and CO2 fluxes in rice-wheat cropping system as affected by crop residues, fertilizer N and legume manure. Biol Fertil Soils 34: 375–389

Aulakh MS, Rennie DA & Paul EA (1983a) Field studies on gaseous nitrogen losses from soils under continuous wheat versus a wheat-fallow rotation. Plant Soil 75: 15–27 Aulakh MS, Rennie DA & Paul EA (1983b) The effect of various clover management practices on gaseous N losses and mineral N accumulation. Can J Soil Sci 63: 593–605 Aulakh MS, Rennie DA & Paul EA (1984a) The influence of plant residues on denitrification rates in conventional and zero tilled soils. Soil Sci Soc Am J 48: 790–794 Aulakh MS, Rennie DA & Paul EA (1984b) Gaseous nitrogen losses from soils under zero-till as compared with conventional-till management systems. J Environ Qual 13: 130–136 Baggs EM, Watson CA & Rees RM (2000) The fate of nitrogen from incorporated cover crop and green manure residues. Nutr Cycl Agroecosyst 56: 153–163 Beauchamp EG, Trevors JT & Paul JW (1989) Carbon sources for bacterial denitrification. Adv Soil Sci 10: 113–142 Bøckman OC & Olfs HW (1998) Fertilizers, agronomy and N2 O. Nutr Cycl Agroecosyst 52: 165–170 Bouwman AF & Sombroek WG (1990) Inputs to climatic change by soils and agriculture related activities: Present status and possible future trends. In: Scharpenseel HW, Schomaker M & Ayoub A (eds) Soils on a Warmer Earth, pp 15–30. Developments in Soil Science, vol 20, Elsevier, Amsterdam Bremner JM, Robbins SG & Blackmer AM (1980) Seasonal variability in emission of nitrous oxide from soil. Geophys Res Lett 7: 641–644 Cai ZC, Laughlin RJ & Stevens RJ (2001) Nitrous oxide and dinitrogen emissions from soil under different water regimes and straw amendment. Chemosphere 42: 113–121 Carran RA,Theobald PW & Evans JP (1995) Emission of nitrous oxide from some grazed pasture soils in New Zealand. Aust J soil Res 33: 341–352 CMDL (2000) Climate Monitoring and Diagnostic Laboratory (CMDL) of the National Oceanographic and Atmospheric Administration, Boulder, CO, USA Cochran VL, Sparrow EB, Schlenter SF & Knight CW (1997) Longterm tillage and crop residue management in the subarctic: fluxes of methane and nitrous oxide. Can J Soil Sci 77: 565–570 Crutzen PJ (1991) Atmospheric chemical processes of the oxides of nitrogen including nitrous oxide. In: Delwiche CC (ed) Denitrification, Nitrification and Atmospheric Nitrous Oxide, pp 17–44. Wiley, New York Duxbury JM, Bouldin DR, Terry RE & Tate RL (1982) Emissions of nitrous oxide from soils. Nature 298: 462–464 Galbally IE, Fraser PJ, Meyer CP & Griffith DWT (1992) Biosphere-atmosphere exchange of trace gases over Australia. In: Gifford RM & Barson MM (eds) Australia’s Renewable Resources Sustainability and Global Change. Bureau of Rural Resources Proceedings No. 14: 117–149. AGPS: Canberra, Australia IPCC (1996) (Intergovernmental Panel on Climate Change/Organization for Economic Cooperation and Development). Guidelines for National Greenhouse Gas Inventories. OECD/OCDE, Paris Jacinthe PA & Dick WA (1997) Soil management and nitrous oxide emissions from cultivated fields in southern Ohio. Soil Till Res 41: 221–235 John PS, Buresh RJ, Prasad R & Pawley RK (1989) Nitrogen gas (N2 +N2 O) flux from urea applied to lowland rice as affected by green manure. Plant Soil 119: 7–13 Khalil MAK & Rasmussen RA (1988) Nitrous oxide: trends and global mass balance over the last 3000 years. Ann Glaciol 10: 73–79

133 Lessard R, Rochette P, Greorich EG, Pattey E & Desjardins RL (1996) Nitrous oxide fluxes from manure-amended soil under maize. J Environ Qual 25: 1371–1377 Li QK (1992) Paddy Soils of China. pp 108–125, 474–490. Science Press, Beijing (in Chinese) MacKenzie AF, Fan MX & Cadrin F (1997) Nitrous oxide emission as affected by tillage, corn-soybean-alfalfa rotations and nitrogen fertilization. Can J Soil Sci 77: 145–152 MacKenzie AF, Fan MX & Cadrin F (1998) Nitrous oxide emission in three years as affected by tillage, corn–soybean–alfalfa rotations, and nitrogen fertilization. J Environ Qual 27: 698–703 McKenney DJ, Wang SW, Drury CF & Findlay WI (1993) Denitrification and mineralization in soil amended with legume, grass, and corn residues. Soil Sci Soc Am J 57: 1013–1020 Mosier A & Kroeze C (2000) Potential impact on the global atmospheric N2 O budget of the increased nitrogen input required to meet future global food demands. Chemosphere-Global Change Sci 2: 465–473 Mosier A, Kroeze C, Nevison C, Oenema O, Seitzinger S & van Cleemput O (1998) Closing the global N2 O budget: nitrous oxide emissions through the agricultural nitrogen cycle: OECD/IPCC/IEA phase II development of IPCC guidelines for national greenhouse gas inventory methodology. Nutr Cycl Agroecosyst 52: 225–248 Mosier AR, Duxbury JM, Freney JR, Heinemeyer O & Minami K

(1996) Nitrous oxide emissions from agricultural fields: Assessment, measurement and mitigation. Plant Soil 181: 95–108 Plant RAJ & Bouman BAM (1999) Modeling nitrogen oxide emissions from current and alternative pastures in Costa Rica. J Environ Qual 28: 866–872 Prinn R, Cunnold D, Rasmussen R, Simmonds P, Alyea F, Crawford A, Fraser P & Rosen R (1990) Atmospheric emissions and trends of nitrous oxide deduced from 10 years of ALE-GAGE data. J Geophys Res 95: 18369–18385 Ragab M, Aldag R, Mohamed S & Mehana T (1994) Denitrification and nitrogen immobilization as affected by organic matter and different forms of nitrogen added to an anaerobic water sediment system. Biol Fert Soils 17: 219–224 Smith KA, McTaggart IP, Dobbie KE & Conen F (1998) Emissions of N2 O from Scottish agricultural soils, as a function of fertilizer N. Nutr Cycl Agroecosyst 52: 123–130 Terry RE, Tate III RL, Duxbury JM (1981) The effect of flooding on nitrous oxide emissions from an organic soil. Soil Sci 132(3): 228–232 Velthof GL, Oenema O, Postma R & Van Beusichem ML (1996) Effects of type and amount of applied nitrogen fertilizer on nitrous oxide fluxes from intensively managed grassland. Nutr Cycl Agroecosyst 46(3): 257–267 Weiss RF (1981) The temporal and spatial distribution of tropospheric nitrous oxide. J Geophys Res 86: 7185–7195