Exploring a suitable nitrogen fertilizer rate to reduce greenhouse gas emissions and ensure rice yields in paddy fields

Science of the Total Environment 565 (2016) 420–426

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Exploring a suitable nitrogen fertilizer rate to reduce greenhouse gas emissions and ensure rice yields in paddy fields Yiming Zhong, Xiaopeng Wang, Jingping Yang ⁎, Xing Zhao, Xinyi Ye Institute of Environmental Science and Technology, College of Environmental and Resource Sciences, Zhejiang University, Hangzhou 310058, China

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Exploiting co-benefits of rice yield and reduction of greenhouse gas emission. • Global warming potential and rice yield increased with nitrogen fertilizer rate up. • Emission peaks of CH4, CO2 and N2O appeared at vegetative and reproductive phase. • 225 kg N/ha rate benefits both rice yields and GWP reduction.

a r t i c l e

i n f o

Article history: Received 16 January 2016 Received in revised form 15 April 2016 Accepted 23 April 2016 Available online xxxx Editor: Ajit Sarmah Keywords: Greenhouse gas Paddy field Nitrogen fertilizer Global warming potential

a b s t r a c t The application rate of nitrogen fertilizer was believed to dramatically influence greenhouse gas (GHG) emissions from paddy fields. Thus, providing a suitable nitrogen fertilization rate to ensure rice yields, reducing GHG emissions and exploring emission behavior are important issues for field management. In this paper, a two year experiment with six rates (0, 75, 150, 225, 300, 375 kg N/ha) of nitrogen fertilizer application was designed to examine GHG emissions by measuring carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O) flux and their cumulative global warming potential (GWP) from paddy fields in Hangzhou, Zhejiang in 2013 and 2014. The results indicated that the GWP and rice yields increased with an increasing application rate of nitrogen fertilizer. Emission peaks of CH4 mainly appeared at the vegetative phase, and emission peaks of CO2, and N2O mainly appeared at reproductive phase of rice growth. The CO2 flux was significantly correlated with soil temperature, while the CH4 flux was influenced by logging water remaining period and N2O flux was significantly associated with nitrogen application rates. This study showed that 225 kg N/ha was a suitable nitrogen fertilizer rate to minimize GHG emissions with low yield-scaled emissions of 3.69 (in 2013) and 2.23 (in 2014) kg CO2-eq/kg rice yield as well as to ensure rice yields remained at a relatively high level of 8.89 t/ha in paddy fields. © 2016 Elsevier B.V. All rights reserved.

1. Introduction

⁎ Corresponding author. E-mail address: [email protected] (J. Yang).

http://dx.doi.org/10.1016/j.scitotenv.2016.04.167 0048-9697/© 2016 Elsevier B.V. All rights reserved.

Anthropogenic GHG emissions in 2010 reached 49 ± 4.5 Gt CO2-eq/y. Though the portion of emissions from Agriculture, Forestry, and Other Land Use (AFOLU) decreased to 24% of the total, the amount of emissions still increased (IPCC, 2014). The concentrations of CO2, CH4 and N2O in

Y. Zhong et al. / Science of the Total Environment 565 (2016) 420–426

2011 were 391 ppm, 1803 ppb, and 324 ppb respectively, exceeding preindustrial levels by approximately 40%, 150%, and 20% (IPCC, 2013). It is reported by the Second National Communication of China that GHG emissions caused by agricultural activities were 819 Mt CO2-eq/y and accounted for 10.97% of China's total GHG emissions. Furthermore, emissions from paddy fields with single-cropping rice accounted for 46.38% of total agricultural GHG emissions (China, 2013). With increasing food demand, GHG emissions from agriculture production become an important source and have to be substantially reduced to minimize the risk of climate change (Godfray et al., 2011). Due to the periodic flooded cycles, the soil environment in paddy fields has unique characteristics, such as the alternation of aerobic and anaerobic conditions. Paddy fields are considered to be a main source of methane and nitrous oxide emissions (Hadi et al., 2010; Harris et al., 1985). CO 2 and CH4 are the end products of anaerobic carbon mineralization (Kirk, 2004). CH4 is produced by methanogens in an environment which the oxygen (O 2 ) and sulfate (SO24 −) are limited. N2O is produced by ammonia-oxidizing bacteria (AOB) and archaea (AOA) via nitrification and denitrification processes in the soil (Kögel-Knabner et al., 2010; Santoro et al., 2011). The application of nitrogen fertilizers in rice cultivation has been commonly adopted to improve the nitrogen availability, with high chemical fertilizer usage, China's nitrogen fertilizer consumption on arable land for permanent crops reached 296.8 kg N/ha, and the rate in paddy is approximately 180 kg N/ha, higher than the Asian average level of 128.1 kg N/ha (FAO, 2013; Ma et al., 2008). The FAO (Food and Agriculture Organization of the United Nations) predicted that the world nitrogen fertilizer demand will be approximately 1.19 × 108 t in 2018 at the annual growth of 1.4%, and China will contribute 18% to the increase. But the use of nitrogen fertilizer may be an important factor that regulates CH4 and N2O emissions (Yao et al., 2012). Previous study declined CH4 emission was due to the combined effects of nitrogen fertilization on production, oxidation, and transportation. As most CH4 is emitted through aerenchyma system of the rice plants, higher tiller numbers (due to an increase in N rate) provide pathway for CH4. In contrast, with high N application the concentration of NH+ 4 increased, which can stimulate CH4 oxidation and lead to a reduction in CH4 emission (K et al., 1989; Liang et al., 2013). Nitrogen has been reported to play an important role in soil C storage. Addition of N increased net C stored in response to additions of straw and root growth, which provides abundant carbon resource for soil respiration releasing CO2 (Snyder et al., 2009). N2O can be generated by both nitrification and denitrification processes. As nitrogen fertilization is the main nitrogen source, nitrogen application rate is main factor influencing N2O emission. A meta-analysis of 78 published studies suggested a general trend of exponentially increasing N2O emissions as N inputs in excess of crop needs (Shcherbak et al., 2014). The N2O response to nitrogen fertilization suggests that agricultural N2O fluxes could be reduced with no or little yield penalty by reducing nitrogen fertilization inputs to the level that just satisfy crop needs (Mcswiney and Robertson, 2005). Meanwhile many crop management practices, such as tillage, the timing of fertilization and resource of nitrogen, can affect GHG emissions, directly by affecting the NO− 3 availability or by modifying the soil microclimate and the cycling of carbon and nitrogen (Snyder et al., 2009). The potential of increasing NO− 3 N in residual soil can be reduced by cutting nitrogen application rates to decrease N2O emission, but this is not considered to be appropriate management because the lack of nitrogen could result in a decrease of soil organic carbon (SOC) and cause a decline in the long-term soil productivity, which would reduce crop yields. Numerous studies have shown that the appropriate use of N fertilizer increases biomass production, and appropriate management practices may lower the risk of increased GHG emissions. An optimum N fertilizer application rate of 225–270 kg N/ha was suggested to achieve good rice productivity in the Taihu Lake region (Deng et al., 2012; Wang,

421

2003). However, we suspect the recommended N fertilizer application rates may be still too high. This study aims to measure and estimate GHG emission potential and emission behavior during the rice growth stage under different nitrogen fertilization rates, to reduce GHG emissions and maintain rice yields at a relatively high level in paddy fields in Hangzhou, China. 2. Materials and methods 2.1. Experimental site and design Field experiments were conducted from June to November 2013 on the experimental farm at Site 1 and from June to November 2014 at Site 2 of the Hangzhou Academy of Agriculture, Hangzhou, Zhejiang China (Table 1). The region has a subtropical monsoon climate with mean annual precipitation of 1454 mm and a mean air temperature of 17.8 °C. Rice cultivar “Hang 43” was planted in a paddy field plot under six nitrogen rates 0 (CK), 75, 150, 225, 300, and 375 kg N/ha by using a randomized block design with three replicates. Each plot size was 3.6 m × 5 m and separated by a high ridge with a plastic film cover to prevent the movement of water and fertilizers between plots. Rice seedlings with five or six fully expanded leaves were transplanted on June 18. Hill spacing was 0.23 m × 0.13 m, with two seedlings per hill, resulting in a plant density of 67.6 plants/m2. Superphosphate (225 kg/ha) and potassium chloride (75 kg/ha) were incorporated into each plot on the day of transplantation; an additional 75 kg/ha of potassium chloride was applied as a top-dressing after transplantation 40 days to prevent K deficiency. Plants received nitrogen in the form of urea, with each N application rate applied in three doses based on the rice growth stages as follows: June 16 (first fertilization, 50%), July 22 (second fertilization, 30%), and August 23 (third fertilization, 20%). 2.2. Measurement of GHG emissions and GWP After rice transplantation and nitrogen fertilization, the opaque plastic static chamber was used weekly to collect the gas from paddy field at 9:00 am from the first fertilization until harvest. An additional sample collection was added during the fertilization week. The static chamber, with a size of 50 cm × 50 cm × 90 cm, was placed over nine hills of rice plants to collect gas. Each sampling was conducted in 10 min intervals for a total of 40 min. Syringes were used to transfer gas from the chamber to 100 mL gas-sampling bag made of aluminum foil. The N2O concentration was measured with a gas chromatograph equipped with an electron capture detector (GC-ECD) operating at 350 °C, while CO2 was reduced to CH4 by methaniser with H2 at 350 °C, and CH4 concentrations were measured by flame ionization detector (GC-FID) operating at 250 °C. Both chromatograph contained a stainless-steel HayeSep Q 80–100 mesh column (outer diameter 3.17 mm), maintained at 60 °C. The increase of GHG concentration in the static chamber was calculated by linear regression. GHG flux was calculated from the increase of GHG concentration using Eq. (1) (Davidson et al., 1998; Zhang et al., 2014).

F¼

dC mPV dC mP  ¼H  dt ART dt RT

ð1Þ

Table 1 The main properties of paddy soil. Year

Site location

Soil type

pH

Organic-C (g/kg soil)

Total-N (g/kg soil)

2013 2014

30.26°N, 120.12°E 30.13°N, 120.16°E

Loam clay Loam clay

5.87 5.57

35.50 12.16

2.75 2.05

422

Y. Zhong et al. / Science of the Total Environment 565 (2016) 420–426

The three GHGs differ in their effectiveness in trapping heat and lifetime in atmosphere. CH4 and N2O are considered as 23 and 296 folds GWP of CO2, respectively (Snyder et al., 2009). GWP ¼ CH4 emission  23 þ N2 O emission  296 þ CO2 emission

ð2Þ

2.3. Soil sampling and analysis

Fig. 1. Variation of atmosphere temperature and logging water in 2013 and 2014.

dC was acquired by the linear regression between the GHG concendt tration and sample time. The parameters H, m, T, P, and R are the height of chamber (m), molecular weight of gas (g/mol), temperature (°C), atmospheric pressure (Pa), and gas constant (R = 8.314 J/mol/K), respectively.

Soil pH was measured after suspending soil in deionized water (1:10 by weight). Soil temperature was measured during gas collection time at 5 cm depth in each plot. Total organic carbon and total nitrogen of soils were determined by dry combustion with an elemental analyzer (Elementar, Germany). Soil samples were collected from a 0 to 10 cm depth in each plot to + measure the nitrate (NO− 3 ) and ammonium (NH4 ) concentrations in 2014. The soils were stored at 4 °C for b7 days before analysis extract− able NH+ 4 and NO3 . Soils were extracted by 25 mL of 1 M KCl solution (Wei et al., 2014) and the concentration was determined by SAN++ Continuous Flow Analyzer (SKALAR, Netherlands). 2.4. Yield-scaled emissions Eq. (3) was used to compare the warming potential emissions per unit of rice yield under different application rates of nitrogen fertilizer. ΣCO 2 ‐ eq is the sum of warming potential (kg CO 2 -eq/ha),

− Fig. 2. Concentrations of soil NH+ 4 -N (A) and NO3 N (B) during the rice growing period in 2014. The error bars represent standard deviations (n = 3). The arrows indicate the time of fertilization.

Y. Zhong et al. / Science of the Total Environment 565 (2016) 420–426

423

Fig. 3. CO2 emission flux under different nitrogen fertilization rates. Fig. 4. CH4 emission flux under different nitrogen fertilization rates.

and CY indicates the rice yield (kg/ha). Ratio ¼

ΣCO2 ‐eq CY

ð3Þ

2.5. Statistical analysis of data A correlation analysis was used to examine the relationship between the GHG emissions and the temperature and the nitrogen fertilization rates. Statistical analyses were performed using SPSS V22.0 (Chicago, IL, USA). GHG emission fluxes are presented as average values with standard deviations using Origin 8.0. 3. Results 3.1. Temperature and logging water As shown in Fig. 1, the daily temperature ranged from 17.4 °C to 37.16 °C in two years, and the temperature in 2013 was higher than in 2014 at the beginning. Depth of logging water depended on rice Table 2 The correlations between GHG emissions and environmental factors. Correlations

CO2 emission

CH4 emission

N2O emission

Atmosphere temperature Soil temperature Logging water Nitrogen treatment Soil NO− 3 Soil NH+ 4

0.046 0.344a −0.172 0.202b −0.288a 0.433a

0.608a 0.527a 0.589a −0.025 −0.345a 0.195

−0.082 0.200 0.096 0.571a −0.038 0.170

a b

Correlation is significant at the 0.01 level (2-tailed). Correlation is significant at the 0.05 level (2-tailed).

growing stage. Before day 80, the logging water was kept at height 1– 5 cm. After day 80, paddy fields were dried to fit local standard cultural practices. The logging water appeared after day 80 in 2014 due to precipitation. According to rice growth habit, the rice cultivation was divided into three phases as vegetative phase (0–56 days), reproductive phase (56–106 days) and ripening phase (1036–141 days). 3.2. The variation of soil nitrogen in paddy field under different N application rates − The concentrations of soil NH+ 4 and NO3 were measured at each single collection of the GHG flux in 2014, as shown in Fig. 2. The concentra− tion of NH+ 4 -N ranged from 6.32 to 78.2 mg/kg. NO3 N ranged from 0.113 to 5.77 mg/kg, with the peak appearing at the end of cultivation because of soil nitrification. A high nitrogen fertilization rate produced + higher NO− 3 N and NH4 -N concentrations.

3.3. GHG flux under different fertilization rates in rice growth Because an opaque static chamber was used to collect GHGs during the day, the photosynthetic activity of the plants was paused at the time of gas collection. Thus, the measured CO2 emission was mainly caused by the respiration from soil microbes and plants. The measured CO2 emission flux ranged from 26.85 mg/m2 h to 2841.81 mg/m2 h during the rice growing period (Fig. 3). The correlation results showed that CO2 emission was significantly correlated with the soil temperature and soil NH+ 4 concentration, with coefficients of 0.344 and 0.433 at the 0.01 level (Table 2). A high nitrogen fertilization rate always got high CO2 emission flux. CH4 is produced by methanogens under an environment where the oxygen (O2) is limited and oxidized by methanotrophic bacteria in the

424

Y. Zhong et al. / Science of the Total Environment 565 (2016) 420–426

Fig. 5. N2O emission flux under different nitrogen fertilizer rates.

surface layer of soil. Though surface soil covered by logging water provides anaerobic environment to accelerate CH4 producing, CH4 can be dissolved in logging water leading to a limited emission, high temperature promoted gases solvent, as the consequence, CH4 emission significantly affected by atmosphere and soil temperature, logging water with coefficient 0.608, 0.527 and 0.589 (Table 2). As shown in Fig. 4, depth of logging water was manipulated during rice growing stage, kept 2–5 cm before 80 days and dried after 80 days to fit local standard cultural practices. CH4 emission is significantly influenced by logging water. CH4 emission peak appeared at vegetative phase then decreased gradually by time. Between 14 day and 22 day, CH4 emission peaked as 32.95 mg/m2 h to 19.62 mg/m2 h for low and high fertilizer rates, respectively. After 70 days, CH4 emission almost declined to zero, and different nitrogen fertilizer rates showed similar emission regulation. The emission flux at vegetative phase in 2013 was higher than that in 2014 due to the soil at Site 1 containing more abundant carbon sources. Fertilization is the main source of nitrogen during rice cultivation. N2O emissions from fertilized fields are affected by some important factors such as climate, the abundance of NO− 3 N in the soil, soil pH and management related factors including the N application rate and crop cultivar. As shown in Fig. 5, the main emission occurred at reproductive phase; higher nitrogen rates always presented high N2O emission flux. N2O emission amount had a significant correlation with nitrogen application rates, with a coefficient of 0.571 (Table 2). 3.4. GHG cumulative emission under different nitrogen fertilization rates The cumulative emission fluxes of CO2, CH4 and N2O are shown in Fig. 6. During the rice growth period under different nitrogen fertilization rates, CO2 emission flux in 2013 was higher than in 2014; in contrast, N2O emission flux in 2014 was higher than in 2013. CH4 emission flux at each

Fig. 6. GHG fluxes under different nitrogen fertilization rates in both years. Different lower case letters indicate significant differences between different nitrogen treatments in 2013. Different capital letters indicate significant differences between different nitrogen treatments in 2014.

N application rate showed no significant differences between the two years except fertilization rate at 0. From the nitrogen application aspect, CO2 and N2O fluxes increased by nitrogen application rate in two years, but CH4 flux showed no significant differences at adjacent nitrogen application rate. The relationship between GHG fluxes and nitrogen fertilization rate fitted from polynomial equation was shown in Table 3. CO2 and N2O fluxes got higher fitting coefficients than CH4. 3.5. Rice fields and yield-scaled emissions under different N application rates The average rice yields under the six fertilization rates ranged from 6.11 t/ha to 9.33 t/ha over two years. As shown in Fig. 7, the yields in 2013 were lower than in 2014, though soil in Site 1 contained higher soil carbon and nitrogen. The main influence factor was climate, as the temperature during vegetative phase in 2014 was higher than in 2013, which provided good condition for rice cultivation. There was no significant increment in rice yields when the N fertilization rate exceeds 225 kg N/ha. The relationship between fertilization rates and rice yields

Y. Zhong et al. / Science of the Total Environment 565 (2016) 420–426

425

Table 3 The GHG flux curves were fitted from polynomial equation. GHG

2013

R2

2014

R2

CO2 CH4 N2 O GWP

y = −0.013 x2 + 8.43 x + 1917.45 y = 8.23^10−5 x2 − 0.057 x + 16.98 y = 3.85^10−5 x 2 + 0.14 x + 8.09 y = −0.13^10−5 x2 + 80.13 x + 21802

0.94 0.75 0.96 0.97

y = −8.39^10−3 x 2 + 6.09 x + 1402.18 y = −9.81^10−5 x 2 + 0.038 x + 7.85 y = 1.52^10−3 x2 − 0.22 x + 43.32 y = −0.056^10−5 x2 + 47.20 x + 18253

0.84 0.47 0.81 0.76

Where y indicates the GHG fluxes (g/m2 for CO2 and CH4 fluxes, mg/m2 for N2O flux and g CO2-eq/m2 for GWP), and x indicates nitrogen application rate (kg/ha).

can be fitted by a nonlinear regression as follows: y ¼ 9:22−2:76  0:99x



 R2 ¼ 0:65; Pb0:05

Where y indicates the rice yields (t/ha), and x indicates nitrogen application rate (kg/ha). The cumulative results for the GWP are shown in Table 4. The results indicated that the contribution from the three gases to the GWP was in the order of CO2 N CH4 N N2O for all treatments. Higher nitrogen treatments led to greater GWP. Taking rice yields into account, the yieldscaled emissions increased with nitrogen application rates from 1.91 to 4.57 kg CO2-eq/kg rice (Table 4). The yield-scaled emissions under 225 kg N/ha treatment were 3.69 and 2.23 in 2013 and 2014, respectively, with an average of 2.96, while a relatively high rice yield was still obtained. 4. Discussion 4.1. Behavior of GHG emissions in paddy field under different N application rates CH4 was generated by logging water that provided an anaerobic environment during vegetative phase, and N 90% of the total CH4 released from rice paddy field was diffusive transport through aerenchyma system of the rice plants (K et al., 1989),. In a previous study, CH4 emission was decreased by reducing the soil redox potential to 36% of the 2-year average under continuous flooding (Minamikawa and Sakai, 2006). In this study, CH4 emission decreased to 0 after vegetative phase and had a correlation coefficient of − 0.345 with soil NO− 3 content, due to NO− 3 N increasing the redox potential. Nitrogen fertilization reduced CH4 emissions because CH4 production was inhibited by nitrite, NO and N2O that accumulated transiently during the reduction of nitrate, which are toxic to methanogens (Klüber and Conrad, 1998; Yao et al., 2012). However, a previous report indicated that nitrogen increases CH4 emission by enhancing methanogen activity and plant growth (Bodelier and Laanbroek, 2004). Because of the amount of proposed underlying mechanisms and the lack of a determination of the methanogen

activity, a mixture of factors affected on CH4 emission caused no significant correlation between the nitrogen fertilization rates. CO2 emission consists of soil respiration and plant respiration. CO2 emission in 2013 was higher than in 2014, and two emission peaks appeared in 2013, while only one appeared in 2014, because the soil at Site 1 contained higher organic carbon as the carbon source of respiration. An increase in CO2 emission was observed during the period from days 60 to 80, when an increase in temperature enhanced rice and soil microbial respiration. Nitrogen uptake by rice strengthened plant respiration, which resulted in a high fertilization rate always generating higher CO2 emission than low N rate. As nitrogen plays an important role in C storage, both by promoting crop dry matter production and by chemically stabilizing C in the soil. Appropriate nitrogen fertilization rates resulted in higher SOC and TN storage in agriculture system (Jagadamma et al., 2007). Some studies have shown that nitrogen fertilization had no significant effect on CO2 emission (Lee et al., 2007; Zhang et al., 2014). However, the CO2 flux increased gradually with the nitrogen application rate (Fig. 6), for which there are three reasons (1) CO2 emission in earlier studies were estimated by the respiration intensity, which was affected by the soil temperature; (2) in this study, a chamber was used to collect CO2 with 9 seedlings involved, and higher nitrogen application rates enhanced plant growth, resulting in stronger respiration; and (3) because an opaque static chamber was used to collect GHGs during the day, the photosynthetic activity of rice plants was paused during gas collection. All these reasons increased CO2 emission. A 6-year field study indicated that seasonal N2O emission varied significantly between years with the nitrogen fertilization rate and source. N2O emission increased with an increasing nitrogen rate (Liang et al., 2013). As shown in Fig. 6, N2O emission was significantly different between years. However, the common trend is that the majority of emission occurred during the reproductive phase (Fig. 5), and N2O emission increased with nitrogen fertilization application rate increased, while its correlation coefficient with nitrogen fertilization rate was 0.571. High nitrogen fertilization rate increased TN in soils, which is the main nitrogen source of N2O. Exceeding nitrogen fertilization and the alternation of aerobic and anaerobic conditions in soil contributed to N2O emission. Though nitrogen contributes to C storage, the positive role of nitrogen fertilization in sequestering C may be offset by N2O emissions (Snyder et al., 2009). Relative N2O emissions from NO− 3 based sources may exceed those from NH+ 4 -based sources, and differences may increase with increasing wetness (Harrison and Webb, 2001). The result of a global meta-analysis showed a general trend of exponentially increasing N2O emission as N inputs increase and exceed crop needs (Shcherbak et al., 2014). According to Fig. 6, a nitrogen treatment of N225 kg N/ha caused significant growth in N2O emission in this study. 4.2. A suitable N rate and appropriate field management for GHGs

Fig. 7. Relationship between rice yield and nitrogen fertilization rates in two years.

An integrated reactive nitrogen budget of China indicated that the N fertilizer use efficiency declined from 1980 to 2010. China continues to enhance crop yields to sustain a growing and ever-wealthier population (Gu et al., 2015). Though some research results have confirmed an appropriate nitrogen fertilization rate and field management could help reduce GHG emissions (Liang et al., 2013; Snyder et al., 2009; Zhang et al., 2014), there was lack of information about the suitable N supply rate after considering the co-benefits of rice production and the

426

Y. Zhong et al. / Science of the Total Environment 565 (2016) 420–426

Table 4 Emissions of GWP (CO2-eq) under different nitrogen fertilization rates. Nitrogen treatment (kg N/ha)

0 75 150 225 300 375

CH4 (g CO2-eq/m2)

CO2 (g CO2-eq/m2)

N2 O (g CO2-eq/m2)

GWP (g CO2-eq/m2)

Yield-scaled emissions (kg CO2/kg rice)

2013

2014

2013

2014

2013

2014

2013

2014

2013

2014

1870.38 2790.23 2927.03 2917.75 3148.63 3263.51

1337.12 1809.79 2141.64 1996.80 2261.44 2556.60

404.63 294.39 261.24 224.49 145.38 232.11

168.00 262.02 261.12 268.80 189.76 213.82

2.29 6.07 8.83 9.10 16.26 16.51

9.19 10.57 35.07 19.98 49.38 51.85

2277.30a 3090.69b 3197.10b 3151.34b 3310.27b 3512.13b

1514.31A 2082.38B 2437.83BC 2285.58BC 2500.58CD 2822.27D

3.82a 4.57c 4.19d 3.69b 3.85d 3.96e

1.91AB 2.40C 2.71B 2.23A 2.73AB 3.03AB

Different lower case letters indicate significant differences between different nitrogen treatments in 2013. Different capital letters indicate significant differences between different nitrogen treatments in 2014.

reduction in GHG emissions in Zhejiang Province. From our field results, we suggest that the appropriate nitrogen fertilization rate for rice production in Hangzhou is approximately 225 kg N/ha, which coincides with the 197–255 kg N/ha range proposed from 3 years of research based on rice yields (Li et al., 2015). Compared to the real average N fertilization of farmers in Zhejiang of 300 kg N/ha, there was a 25% reduction in nitrogen fertilizer use, a 187.00 g CO2-eq/m2 emission reduction, and nearly 0.33 kg CO2-eq/kg rice reduction in the yield-scaled emissions during the rice production period (Table 4). Observations of N2O and CH4 emissions in response to water management suggested that irrigation control will decrease CH4 emission by 81.8%, increase N2O emission by 135.4%, and decrease the integrated GWP by 27.3% (Hou et al., 2012). In our study, the field was dried during days 35–50 to increase effective rice tillering, while the CH4 emission rate decreased by 67.05% immediately (Fig. 4). Therefore, we suggest controlling water irrigation in the vegetative phase to limit CH4 emission. Because N2O emission in this study occurred mainly during the reproductive phase when the logging water was deficient, the supplement of water at reproductive phase for a period may help inhibit N2O emission. 5. Conclusions GHG emissions during rice cultivation are significantly affected by the application rates of nitrogen fertilizer, the soil temperature and logging water. Appropriate fertilizer application rates result in low nitrogen consumption, low GHG emissions and relatively high rice yields. Emission peaks of CH4 mainly appeared at the vegetative phase, and emission peaks of CO2, and N2O mainly appeared at reproductive phase of rice growth. According to results of this two year study, the suitable N fertilizer in Hangzhou in a loam paddy is approximately 225 kg N/ha, which had low yield-scaled emissions of 3.69 and 2.23 kg CO2-eq/kg rice yields in two years. Acknowledgements We thank Dr. Prof. Chunlong Zhang from University of Houston for helpful comments on edits for the manuscript. This study was supported by the National Natural Science Foundation of China (no. 61174089). References Bodelier, P.L.E., Laanbroek, H.J., 2004. Nitrogen as a regulatory factor of methane oxidation in soils and sediments. FEMS Microbiol. Ecol. 47, 265–277. China, P.R., 2013. Second National Communication on Climate Change of the People's Republic of China, Beijing, China. Davidson, E.A., Belk, E., Boone, R.D., 1998. Soil water content and temperature as independent or confounded factors controlling soil respiration in a temperate mixed hardwood forest. Glob. Chang. Biol. 4, 217–227. Deng, M., Shi, X., Tian, Y., Yin, B., Zhang, S., Zhu, Z., et al., 2012. Optimizing nitrogen fertilizer application for rice production in the Taihu Lake Region, China. Pedosphere 22, 48–57. FAO, 2013. FAO Statistical Yearbook 2013 Part 1-The Setting, Rome.

Godfray, H.C.J., Pretty, J., Thomas, S.M., Warham, E.J., Beddington, J.R., 2011. Linking policy on climate and food. Science 331, 1013. Gu, B., Ju, X., Chang, J., Ge, Y., Vitousek, P.M., 2015. Integrated reactive nitrogen budgets and future trends in China. Proc. Natl. Acad. Sci. 112, 8792–8797. Hadi, A., Inubushi, K., Yagi, K., 2010. Effect of water management on greenhouse gas emissions and microbial properties of paddy soils in Japan and Indonesia. Paddy Water Environ. 8, 319–324. Harris, R.C., Gorham, E., Sebacher, D.I., Bartlett, K.B., Flebbe, P.A., 1985. Methane flux from northern peatland. Nature 652–654. Harrison, R., Webb, J., 2001. A review of the effect of N fertilizer type on gaseous emissions. Adv. Agron. 73, 65–108. Hou, H., Peng, S., Xu, J., Yang, S., Mao, Z., 2012. Seasonal variations of CH4 and N2O emissions in response to water management of paddy fields located in Southeast China. Chemosphere 89, 884–892. IPCC, 2013. Climate change 2013 the physical science basis. Summary for policymakers. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. IPCC, 2014. Climate Change 2014 Synthesis Report Summary for Policymakers Chapter. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Jagadamma, S., Lal, R., Hoeft, R.G., Nafziger, E.D., Adee, E.A., 2007. Nitrogen fertilization and cropping systems effects on soil organic carbon and total nitrogen pools under chisel-plow tillage in Illinois. Soil Tillage Res. 95, 348–356. K, I., K, H., S, M., M, U., H, W., 1989. Methane emission from the flooded paddy soil to the atmosphere through rice plant. Jpn. J. Soil Sci. Plant Nutr. 318–324. Kirk, G., 2004. The Biogeochemistry of Submerged Soils. Wiley. Klüber, H.D., Conrad, R., 1998. Effects of nitrate, nitrite, NO and N2O on methanogenesis and other redox processes in anoxic rice field soil. FEMS Microbiol. Ecol. 25, 301–318. Kögel-Knabner, I., Amelung, W., Cao, Z., Fiedler, S., Frenzel, P., Jahn, R., et al., 2010. Biogeochemistry of paddy soils. Geoderma 157, 1–14. Lee, D.K., Doolittle, J.J., Owens, V.N., 2007. Soil carbon dioxide fluxes in established switchgrass land managed for biomass production. Soil Biol. Biochem. 39, 178–186. Li, Y., Chen, Y., Wu, C., Tang, X., Ji, X., 2015. Determination of optimum nitrogen application rates in Zhejiang Province, China, based on rice yields and ecological security. J. Integr. Agric. 14, 2426–2433. Liang, X.Q., Li, H., Wang, S.X., Ye, Y.S., Ji, Y.J., Tian, G.M., et al., 2013. Nitrogen management to reduce yield-scaled global warming potential in rice. Field Crop Res. 146, 66–74. Ma, W., Li, J., Ma, L., Wang, F., Sisák, I., Cushman, G., et al., 2008. Nitrogen flow and use efficiency in production and utilization of wheat, rice, and maize in China. Agric. Syst. 99, 53–63. Mcswiney, C.P., Robertson, G.P., 2005. Nonlinear response of N2O flux to incremental fertilizer addition in a continuous maize (Zea mays L.) cropping system. Glob. Chang. Biol. 11, 1712–1719. Minamikawa, K., Sakai, N., 2006. The practical use of water management based on soil redox potential for decreasing methane emission from a paddy field in Japan. Agric. Ecosyst. Environ. 116, 181–188. Santoro, A.E., Buchwald, C., Mcilvin, M.R., Casciotti, K.L., 2011. Isotopic signature of N2O produced by marine ammonia-oxidizing archaea. Science 333, 1282–1285. Shcherbak, I., Millar, N., Robertson, G.P., 2014. Global metaanalysis of the nonlinear response of soil nitrous oxide (N2O) emissions to fertilizer nitrogen. Proc. Natl. Acad. Sci. 111, 9199–9204. Snyder, C.S., Bruulsema, T.W., Jensen, T.L., Fixen, P.E., 2009. Review of greenhouse gas emissions from crop production systems and fertilizer management effects. Agric. Ecosyst. Environ. 133, 247–266. Wang, D.J., 2003. Optimum nitrogen rate for a high productive rice-wheat system and its impact on the groundwater in the Taihu lake area. Acta Pedol. Sin. 40, 426–432. Wei, W., Isobe, K., Shiratori, Y., Nishizawa, T., Ohte, N., Otsuka, S., et al., 2014. N2O emission from cropland field soil through fungal denitrification after surface applications of organic fertilizer. Soil Biol. Biochem. 69, 157–167. Yao, Z., Zheng, X., Dong, H., Wang, R., Mei, B., Zhu, J., 2012. A 3-year record of N2O and CH4 emissions from a sandy loam paddy during rice seasons as affected by different nitrogen application rates. Agric. Ecosyst. Environ. 152, 1–9. Zhang, X., Yin, S., Li, Y., Zhuang, H., Li, C., Liu, C., 2014. Comparison of greenhouse gas emissions from rice paddy fields under different nitrogen fertilization loads in Chongming Island, Eastern China. Sci. Total Environ. 472, 381–388.