论文部分内容阅读
Abstract It is of important referential values for the further understanding of the effects of fertilization on greenhouse gas emissions and the effects of winter green manure on soil carbon pool to study the effects of fertilization on the greenhouse gas emissions and soil carbon pool during the growing season of winter Chinese milk vetch in the process of rice cultivation. This study investigated the effects of nitrogen application in late rice season on the yield of the succeeding Chinese milk vetch and greenhouse gas emissions as well as the soil carbon pool characteristics after the winter planting of Chinese milk vetch with the winter idling of no nitrogen application as the control. The results showed that the yield of Chinese milk vetch was the highest under the nitrogen application of 225 kg/hm 2 in the late rice season， reaching up to 18 388.97 kg/hm 2， which was significantly different from other treatments （P<0.05）. Nitrogen application in late rice season increased the emissions of N2O， CH4， CO2 and global warming potential （GWP） in the growing season of Chinese milk vetch. Compared with the winter idling treatment， winter planting of Chinese milk vetch significantly increased the soil organic carbon and soil carbon pool management index. The yield of Chinese milk vetch was significantly positively correlated with N2O and CH4 emissions （P<0.05）， while it presented extremely significant positive correlations with CO2 emissions， GWP， active organic carbon， and carbon pool management index （P<0.01）. Nitrogen application in the late rice season increased the emissions of N2O， CH4， CO2， and enhanced the greenhouse gas emission potential during the growing season of Chinese milk vetch. Therefore， without reducing the yield of rice， reducing the amount of nitrogen fertilizer in rice could reduce the greenhouse gas emissions in the growing season of succeeding Chinese milk vetch.
Key words Nitrogen application; Chinese milk vetch; Greenhouse gas emission; Soil carbon pool
Farmland soil carbon pool is an important indicator to measure soil fertility. Cultivation methods like winter green manure incorporation， rice straw incorporation and combined application of green manure with nitrogen fertilizer can directly carry exogenous organic matters to soil， thereby increasing soil active organic carbon content[1] and improving soil carbon pool management index[2-6]. Moreover， soil organic carbon content is closely related to the global "greenhouse effect"[7]， and mulching and turning green manure crops can significantly increase soil organic carbon content， thus affecting greenhouse gas emissions[8]. Chinese milk vetch is one of the important green manure crops in the rice planting region in south China. It has strong nitrogenfixing capacity. After incorporating into field， Chinese milk vetch can replace some of the chemical nitrogen fertilizers needed for the growth of the succeeding rice， and reduce the amount of nitrogen fertilizer， thereby effectively improving soil physical and chemical properties， increasing soil organic matter content and microbial content， increasing crop yield， and improving rice quality[9-13]. At present， there are many studies on greenhouse gas emissions during the growth period of rice[14-16]， but there are few studies on greenhouse gas emissions during the growing season of winter green manure[17-18]， while for winter rice fertilization， the growth of winter green manure， and there is still no report on the effects of fertilization to late rice on greenhouse gas emissions during the growing season of winter green manure. It is of important significance for the energy conservation and emission reduction of paddy fields to study the effects of nitrogen application rate in rice cultivation on greenhouse gas emissions and soil carbon pool during the growing season of Chinese milk vetch， as well as the mutual relations between them. In order to better evaluate the ecological effects of planting Chinese milk vetch in winter fallowing field， the effects of nitrogen application to late rice on the yield of fresh Chinese milk vetch， greenhouse gas emission characteristics during the growing season and soil carbon pool features after planting winter Chinese milk vetch were explored with winter fallowing treatment as the control， so as to better develop and utilize the resources of winter fallowing land and Chinese milk vetch resources， thereby providing a theoretical basis for energy saving and emission reduction in paddy fields. Materials and Methods
Test site overview
The test of nitrogen fertilization to Chinese milk vetch began in October 2011. The greenhouse gas collection time in the growing season was from November 2015 to March 2016. The test was conducted in the experimental field of Yujiang Agricultural Science Research Institute， Yingtan City， Jiangxi Province （116°41′-117°09′ E， 28°04′-28°37′N）. The test site had subtropical monsoon humid climate with the annual average temperature of 17.6 ℃， average accumulated temperature above ≥0 ℃ of 6 586.4 ℃， annual precipitation of about 1 741 mm， frostfree period of 258 d， and total annual solar radiation of 454.27 kJ/cm 2， and annual average wind speed of 1.0-3.8 m/s. The soil was mostly siltation soil. At the beginning of the test， the surface soil （0-15 cm） had an organic matter content of 20.65 g/kg， total nitrogen content of 1.85 g/kg， total phosphorus content of 0.48 g/kg， alkali nitrogen content of 151.00 mg/kg， available phosphorus content of 59.76 mg/kg， available potassium content of 38 mg/kg， and pH value of 5.59.
Test design
A total of 5 treatments were designed in the test， namely， ① Treatment A （CK） of winter fallowing， in which the nitrogen application rate in the late rice season was 0 kg/hm 2; ② Treatment B of winter planting Chinese milk vetch， in which the nitrogen application rate in the late rice season was 0 kg/hm 2; ③ Treatment C of winter planting Chinese milk vetch， in which the nitrogen application rate in the late rice season was 90 kg/hm 2; ④ Treatment D of winter planting Chinese milk vetch， in which the nitrogen application rate in the late rice season was 150 kg/hm 2; ⑤ Treatment E of winter planting Chinese milk vetch， in which the nitrogen application rate in the late rice season was 225 kg/hm 2. Each treatment had 3 repetitions， following the completely randomized block design. The application rates of phosphate and potassium fertilizers were the same in all the late rice treatments， in which the application rate of phosphate fertilizer was 60 kg/hm 2 and the application rate of potassium fertilizer was 75 kg/hm 2. The application rate of chemical fertilizers were calculated using urea with 46% of N， calcium magnesium phosphate with 12% of P2O5，and potassium chloride with 60% of K2O， and no treatment had the straws returned to fields after the harvest of late rice. The test plot was 11 m long and 6 m wide with an area of 66 m 2， and guarding rows were set on both sides with the width of 1 m. The test Chinese milk vetch variety was Yujiang Dayezi， which was sown on October 1， 2015 with the sowing amount of 27.5 kg/hm 2. The turning time was on March 28， 2016. Sample collection and determination
Yield measurement of Chinese milk vetch
In the flowering period of Chinese milk vetch， the 5point method was adopted to measure its fresh weight by taking 1 m 2 from each point. The total yield of fresh grass was the plot area multiplying the average yield of the 5 points.
Greenhouse gas collection and determination
The greenhouse gas was collected using static box method. The static box was 50 cm in length， width and height. The inside was made of stainless steel plate， and the outside was packed with sponge and aluminum foil insulation board of 0.5 cm thick to avoid the temperature in box rise too quickly because of the sun. There was a 12 V small fan equipped at the top of the box to prevent the gas in the box from being uneven. There was a pumping hole in the middle of the side of the box， which was controlled by a threeway valve control switch. Each plot was fixed with a sampling base， and there was a groove of 5 cm deep on the base which was used to seal with water when measuring. During the growing season of Chinese milk vetch， samples were taken every 10-15 d[19]， and the daily average temperature was recorded during the sampling period （Fig. 1）. The collection time was 08：00-11：00. Samples were collected at a time interval of 0， 10， 20， 30 min， during which a 50 mL injector was used to pump the gas inside the box for 5-10 times to fully mix the gas， and then 50 mL of gas was pumped and quickly taken back to the laboratory for analysis after being stored in a vacuum sampling bag. The concentration of N2O， CH4 and CO2 was determined by Agilent 7890 B gas chromatography. The detector for measuring CH4 and CO2 was FID with the detection temperature of 300 ℃ and column temperature of 60 ℃， and the carrier gas was 99.99% highpurity nitrogen gas with a flow rate of 30 mL/min. The detector for measuring N2O was ECD with the detection temperature of 300 ℃ and column temperature of 60 ℃， and the carrier gas was 99.99% highpurity argon/methane gas （95% argon + 5% methane） at a flow rate of 40 mL/min. Gas emission flux calculation formula was as follows：
F = ρ × h × dC/dt × 273/（ 273 + T） .
Where： F is the gas emission flux， mg/（m 2·h） or μg/（ m 2·h） ; ρ is the density of the gas under standard conditions， kg/m 3; h is the net height of the sampling box， m; dC/dt is the rate of change of the concentration of gas in the sampling tank per unit time; T is the average temperature in the sampling box during sampling， ℃; 273 is the constant of the gas equation. Global warming potential （GWP） converted the warming potential of the total seasonal emissions of various greenhouse gases into CO2 equivalents， and CH4 and N2O were 25 times and 298 times that of CO2 on the 100year scale[20]. The calculation formula was as follows：
GWP = 25（CH4） + 298（ N2O） + CO2.
Determination of soil organic carbon pool index
Before turning Chinese milk vetch into soil， soil samples were taken from each plot according to the 5point sampling method after fully mixing. Some soil was naturally dried for the determination of total soil organic carbon， and the other part of soil was stored in a refrigerator at 4 ℃ for the determination of active organic carbon. For the reference paddy field， the soil total organic carbon mass fraction was 16.67 g/kg， and the active organic carbon mass fraction was 2.11 g/kg. Soil organic carbon （SOC） was determined by potassium dichromateconcentrated sulfuric acid external heating method[21]; active organic carbon （AOC） was determined by 333 mmol/L potassium permanganate oxidation method[22]. The following were the calculation formulas for soil active organic carbon pool and carbon pool management index：
Carbon pool index = Soil organic carbon mass fraction （g/kg）/Referential cropland soil organic carbon mass fraction （g/kg）;
Steady carbon = Soil organic carbonActive organic carbon;
Carbon pool activity = Soil active organic carbon mass fraction （g/kg）/Steady carbon mass fraction （g/kg）;
Carbon pool activity index = Sample carbon pool activity/Referential soil carbon pool activity;
Carbon pool management index = Carbon pool index × Carbon pool activity index × 1 000.
Data processing
Excel 2007 and SPSS 17. 0 were used for statistical analysis of the test data， LSD was used to test the significance of difference between the sample mean values， and Origin was used for mapping.
Results and analysis
Effect of nitrogen application in late rice season on the yield of fresh Chinese milk vetch
As shown in Fig. 2， different nitrogen application rates in late rice season had significant effects on the yield of fresh Chinese milk vetch. With the increase of nitrogen application rate， the yield of Chinese milk vetch increased continuously. The yield of fresh Chinese milk vetch reached the highest in treatment E of up to 18 388.97 kg/hm 2， which was 13.94% higher than that of treatment B， and the difference between the 2 treatments reached the significant level （P<0.05）. The yield of treatment D came to the second， which was 9.98% higher than that of treatment B， and the difference between the two was significant （P<0.05）. However， there was no significant difference between treatment D and treatment C. therefore， compared with no nitrogen application in late rice season， the application of nitrogen in the late rice season could significantly increase the yield of fresh winter Chinese milk vetch. Effects of nitrogen application in late rice season on greenhouse gas emissions during the growing season of Chinese milk vetch
Nitrogen application in late rice season and winter Chinese milk vetch could affect N2O emissions from paddy fields. As shown in Fig. 3， the emissions of N2O from paddy fields were small under different treatments from after sowing Chinese milk vetch to the beginning of December 2015， and N2O emissions increased gradually after December， reaching the highest in midJanuary， 2016. During that period， the emissions of treatments C， D， E were 21.80%， 27.68% and 24.40% higher than those of treatment A， and 6.90%， 14.71%， 12.17% higher than those of treatment B. Afterwards， the N2O emission flux of each treatment gradually decreased until tilling.
Yanqin MA et al. Effects of Nitrogen Application Rate to Late Rice on Greenhouse Gas Emissions and Soil Carbon Pool During the Growing Season of Winter Chinese Milk vetch
As shown in Fig. 4， CH4 emissions were small in the early growth stage because of the low temperature and slow growth of Chinese milk vetch， which were even less than 0， presenting as the absorption of CH4 by soil. From late December， with the gradual increase of temperature and accelerating growth of Chinese milk vetch， the biomass of the aboveground part increased gradually， and the CH4 emission flux of each treated paddy field gradually increased. In late February of the following year， there was a small emission peak of CH4 in each treated rice field， when the CH4 emission flux was the highest in treatment E， reaching 0.40 mg/（m 2·h）， an increase of 110.53% from treatment A. The difference between the two was significant （P<0.05）.
As shown in Fig. 5， nitrogen application in late rice season had a great influence on CO2 emissions. With the growth of Chinese milk vetch， the CO2 emissions in each treatment increased， especially after midFebruary， the CO2 emission flux of each treatment increased rapidly. By the end of February， there was a small peak in each treatment. Compared with treatment A， the increase of CO2 emission flux was in the range of 124.36%-277.09% for treatments B， C， D， and E in this period. Then CO2 emission flux gradually decreased， reaching the maximum before turning over at the end of March.
Effects of nitrogen application in late rice season on total greenhouse gas emissions and global warming potential in the growing season of Chinese milk vetch As shown in Table 1， the total greenhouse gas emissions during the growing season of Chinese milk vetch under treatments C， D， E were significantly higher than those under treatment A of winter idling（P<0.05）. The total accumulated emissions of N2O and CO2 and global warming potential were the highest in treatment E， which showed significant differences from those of treatment B with no nitrogen application （P<0.05）， namely 11.54%， 28.19%， 26.00% higher than those of treatment B. Among the 3 treatments of C， D， E， the emissions of N2O and CO2 and global warming potential were the lowest in treatment C， which were 8.28%， 7.63%， 8.13% lower than those of treatment E （P<0.05）. Treatment showed significant differences in N2O emissions from treatment E （P<0.05）， but there was no significant difference in the total accumulated emissions of CH4 and CO2 and global warming potentials from treatment E （P>0.05）. Therefore， nitrogen application in late rice season could significantly affect the greenhouse gas emissions during the growing season of the succeeding Chinese milk vetch. Moreover， with the increase of nitrogen application rate， the emissions of N2O and CO2 during the growing season of Chinese milk vetch also gradually increased， and the global warming potential also became stronger.
Effects of winter Chinese milk vetch on soil carbon pool
As shown in Table 2， there was no significant difference in active organic carbon and soil carbon pool activity among all the treatments （P>0.05）. Among all treatments， treatment D had the highest total organic carbon content， soluble organic carbon content and carbon pool index， all of which showed significant differences from the control treatment A （P<0.05）， increasing by 21.27%， 32.26%， 11.36%， respectively. Soil active organic carbon content and carbon pool management index of treatment C were the highest， which were 12.23% and 11.99% higher than those of treatment A. The soil carbon pool management indices of treatments B， C， D， E were all significantly higher than that of treatment A with the increase ranging from 9.54% to 11.99%. However， there was no significant difference between treatments B， C， D， E （P>0.05）. Therefore， compared with winter idling treatment， winter planting Chinese milk vetch could significantly increase soil total organic carbon content， soluble organic carbon and carbon pool index， and the soil carbon pool management index was relatively higher with the nitrogen application rate of 90 and 150 kg/hm 2 in late rice season. Correlation analysis between Chinese milk vetch yield， carbon pool management index and greenhouse gas emissions
As shown in Table 3， the nitrogen application rate was significantly correlated with the N2O emissions （P<0.05）， indicating that the input of nitrogen fertilizer could increase the N2O emissions from the paddy field. The yield of Chinese milk vetch showed a significant positive correlation with N2O and CH4 emissions （P<0.05）， and extremely significant positive correlations with active organic carbon， carbon pool management index， CO2 emissions， and global warming potential （P<0.01）， indicating that with the increase of fresh Chinese milk vetch yield， the soil active organic carbon content and carbon pool management index were gradually increased， the greenhouse gas emissions also increased during the growing season of Chinese milk vetch， and the global warming potential also became stronger. The soil carbon pool management index showed extremely significant positive correlations with and CO2 emissions and global warming potential （P<0.01）， and significant positive correlations with N2O and CH4 emissions （P<0.05）. Therefore， changes in soil carbon content could affect greenhouse gas emissions in paddy fields.
Discussion
Fertilization and winter Chinese milk vetch affecting greenhouse gas emissions from paddy fields
Nitrogen fertilizer plays an important role in increasing grain yield， and it also produces a large amount of greenhouse gases such as N2O， CH4 and CO2. Studies have shown that the application of chemical nitrogen fertilizer is the main factor promoting N2O emissions from farmland， and the decisive degree of nitrate nitrogen on farmland N2O emissions is 65%[23]. The results of this study show that the total amount of N2O emissions under the treatments of winter planting Chinese milk vetch is greater than those of winter idling treatment， which is similar to the results of previous studies[17-18， 24]. Moreover， the results also show that the N2O emission flux under each treatment is higher in the late growth stage than that in the early stage， which is consistent with the results of OHara et al.[25]. The main reason is that the soil microbial activities gradually increase with the temperature， and because of the welldeveloped root system of Chinese milk vetch， the biomass of the aboveground part also increases with the increase of nitrogen application rate in the late rice season， and its physiological activity increases， resulting in the increase of N2O emissions. This view is also confirmed by a significant positive correlation between N2O emissions and the yield of fresh Chinese milk vetch. This study shows that CH4 emissions were low before midJanuary 2016， and even negative emissions appeared at certain times， while CH4 emission flux increased significantly after midDecember， which is consistent with the results of Tang et al.[17， 26]. The main reason is that CH4 emissions from paddy fields are closely related to soil water content[27] and temperature[28-29]. The temperature gradually increases after late December， and the suitable temperature， increased precipitation and good soil environment are conducive to the growth of methanogens， leading to a small peak of at least once in the CH4 emissions during the growing season of the milk vetch. The temperature drops after midJanuary， and the CH4 emissions also gradually decrease. After February， as the temperature increases， the CH4 emissions show a peak again. Winter Chinese milk vetch improving farmland soil carbon management index
The application of organic fertilizer or the combined application of organic fertilizer with inorganic fertilizer directly inputs exogenous organic matters into the soil， which can significantly increase the organic carbon content[30-31]. It has found that soil carbon pool management indices of the winter idling control are all the lowest， and the winter planting of Chinese milk vetch significantly increases the soil organic carbon and soil carbon pool management index， mainly because the notillage and direct seeding of Chinese milk vetch， on the one hand， can increase soil organic carbon storage， and on the other hand， it can increase the ground coverage， reduce soil disturbance times and soil nutrients loss， lower the erosion of wind and rain to soil， and supplement some lost carbon to the soil， which in turn enhances soil carbon sequestration capacity[33]， increases soil carbon pool activity， carbon pool activity index， and carbon stock index. At the same time， the test results also show that the content of soil active organic carbon under winter Chinese milk vetch treatments is significantly higher than that in the winter idling control， and the correlation analysis results also show that there is a significant positive correlation between the yield of fresh Chinese milk vetch and soil active organic carbon. The main reason is that the yield of fresh Chinese milk vetch gradually increases with the increase of nitrogen application rate under different nitrogen application treatments in late rice season， which improves soil active organic carbon content.
Close relations between paddy field greenhouse gas emissions and soil carbon pool
As one of the most active carbon pools in the surface ecosystem， soil carbon pool is an important release source of greenhouse gases such as N2O， CH4 and CO2， and an important sink for the greenhouse gases[34]. The increase of soil organic carbon storage in paddy fields is of great significance for decreasing soil greenhouse gas emissions， reducing atmospheric greenhouse gas concentrations， and alleviating current global warming problems. The results of this study show that there is a significant positive correlation between N2O and CH4 emissions and soil carbon pool management index （P<0.05）， and CO2 emissions and global warming potential show extremely significant positive correlations with soil carbon pool management index （P<0.01）. The reason is that notillage and direct seeding of Chinese milk vetch reduce the number of tillage in the paddy field， and the soil mineralization rate is increased during the growth of Chinese milk vetch. Moreover， the soil can be kept warm during the growth period of Chinese milk vetch. In this way， it accelerates the decay of soil organic matter， increases soil carbon pool management index， thereby increasing the N2O， CH4， CO2 emissions and greenhouse gas emission potential during the growing season of Chinese milk vetch. The global warming potential of Chinese milk vetch season shows extremely positive correlations with the yield of Chinese milk vetch and CH4 and CO2 emissions （P<0.01）， and it shows a significant positive correlation with N2O emissions （P<0.05）， which are similar to the results of Mosier et al.[35-36]. The reason is that the soil carbon sequestration capacity and carbon fixation amount are higher at high yield of winter Chinese milk vetch[32-33]， and the high carbon reserves enhance greenhouse gas emissions from farmland soils. Although the winter planting of Chinese milk vetch increases the greenhouse gas emissions of paddy fields， returning Chinese milk vetch to field can substitute some nitrogen fertilizers， which can reduce the greenhouse gas emissions from paddy fields to some extent， offsetting some of the greenhouse gas emission increase effect[37]. Conclusion
Nitrogen application in late rice season has a significant effect on the yield of the succeeding Chinese milk vetch， and the yield of Chinese milk vetch increases with the increase of nitrogen application rate. Winter planting of Chinese milk vetch can significantly increase soil organic carbon content and soil carbon pool management index. Nitrogen application in late rice season can increase the N2O， CH4， and CO2 emissions during the growing season of Chinese milk vetch， and enhance the greenhouse gas emission potential. Therefore， without reducing the yield of rice， reducing the amount of nitrogen fertilizer in the rice season can reduce the greenhouse gas emissions during the growing season of the succeeding Chinese milk vetch.
References
[1] ZHANG GL， ZHAO JN， SONG XL， et al. Effects of fertilization on soil organic carbon and carbon pool management index[J]. Plant Nutrition and Fertilizer Science， 2012， 18（ 2） ： 359-365.
[2] XU MG， YU R， SUN XF， et al. Effects of longterm fertilization on labile organic matter and carbon pool management index （CMI） of the typical soils of China[J]. Plant Nutrition and Fertilizer Science， 2006， 12（ 4） ： 459-465.
[3] WU HF， ZENG YH， PAN XH， et al. Effect of rice straw incorporation on rice yield and carbon pool management index under the application of farm mechanization[J]. Acta Agriculturae Universitatis Jiangxiensis （Natural Sciences Edition）， 2011， 33（ 5） ： 835-839， 879.
[4] CHEN SH， ZHU ZL， LIU DH， et al. Influence of straw mulching with notill on soil nutrients and carbon pool management index[J]. Plant Nutrition and Fertilizer Science， 2008， 14（4）： 806-809.
[5] WANG J， ZHU P， ZHANG N， et al. Effect of fertilization on soil active C and C pool management index of black soil[J]. Chinese Journal of Soil Science， 2003， 34（5）： 394-397.
[6] YU WT， ZHAO X， MA Q， et al. Effects of longterm fertilization on available carbon pool and carbon pool management index in an aquic brown soil[J]. Chinese Journal of Soil Science， 2008， 39（3）： 539-544.
[7] WEST TO， MARLAND G. A synthesis of carbon sequestration， carbon emission， and net carbon flux in agriculture： comparing tillage practices in the United States[J]. Agriculture Ecosystems & Environment， 2002， 91（1 /2 /3）： 217-232.
[8] TANG HM， XIAO XP， TANG WG， et al. Effects of straw recycling of winter covering crop on methane and nitrous oxide emissions in paddy field[J]. Acta Agronomica Sinica， 2011， 37（9）： 1666-1675.
Key words Nitrogen application; Chinese milk vetch; Greenhouse gas emission; Soil carbon pool
Farmland soil carbon pool is an important indicator to measure soil fertility. Cultivation methods like winter green manure incorporation， rice straw incorporation and combined application of green manure with nitrogen fertilizer can directly carry exogenous organic matters to soil， thereby increasing soil active organic carbon content[1] and improving soil carbon pool management index[2-6]. Moreover， soil organic carbon content is closely related to the global "greenhouse effect"[7]， and mulching and turning green manure crops can significantly increase soil organic carbon content， thus affecting greenhouse gas emissions[8]. Chinese milk vetch is one of the important green manure crops in the rice planting region in south China. It has strong nitrogenfixing capacity. After incorporating into field， Chinese milk vetch can replace some of the chemical nitrogen fertilizers needed for the growth of the succeeding rice， and reduce the amount of nitrogen fertilizer， thereby effectively improving soil physical and chemical properties， increasing soil organic matter content and microbial content， increasing crop yield， and improving rice quality[9-13]. At present， there are many studies on greenhouse gas emissions during the growth period of rice[14-16]， but there are few studies on greenhouse gas emissions during the growing season of winter green manure[17-18]， while for winter rice fertilization， the growth of winter green manure， and there is still no report on the effects of fertilization to late rice on greenhouse gas emissions during the growing season of winter green manure. It is of important significance for the energy conservation and emission reduction of paddy fields to study the effects of nitrogen application rate in rice cultivation on greenhouse gas emissions and soil carbon pool during the growing season of Chinese milk vetch， as well as the mutual relations between them. In order to better evaluate the ecological effects of planting Chinese milk vetch in winter fallowing field， the effects of nitrogen application to late rice on the yield of fresh Chinese milk vetch， greenhouse gas emission characteristics during the growing season and soil carbon pool features after planting winter Chinese milk vetch were explored with winter fallowing treatment as the control， so as to better develop and utilize the resources of winter fallowing land and Chinese milk vetch resources， thereby providing a theoretical basis for energy saving and emission reduction in paddy fields. Materials and Methods
Test site overview
The test of nitrogen fertilization to Chinese milk vetch began in October 2011. The greenhouse gas collection time in the growing season was from November 2015 to March 2016. The test was conducted in the experimental field of Yujiang Agricultural Science Research Institute， Yingtan City， Jiangxi Province （116°41′-117°09′ E， 28°04′-28°37′N）. The test site had subtropical monsoon humid climate with the annual average temperature of 17.6 ℃， average accumulated temperature above ≥0 ℃ of 6 586.4 ℃， annual precipitation of about 1 741 mm， frostfree period of 258 d， and total annual solar radiation of 454.27 kJ/cm 2， and annual average wind speed of 1.0-3.8 m/s. The soil was mostly siltation soil. At the beginning of the test， the surface soil （0-15 cm） had an organic matter content of 20.65 g/kg， total nitrogen content of 1.85 g/kg， total phosphorus content of 0.48 g/kg， alkali nitrogen content of 151.00 mg/kg， available phosphorus content of 59.76 mg/kg， available potassium content of 38 mg/kg， and pH value of 5.59.
Test design
A total of 5 treatments were designed in the test， namely， ① Treatment A （CK） of winter fallowing， in which the nitrogen application rate in the late rice season was 0 kg/hm 2; ② Treatment B of winter planting Chinese milk vetch， in which the nitrogen application rate in the late rice season was 0 kg/hm 2; ③ Treatment C of winter planting Chinese milk vetch， in which the nitrogen application rate in the late rice season was 90 kg/hm 2; ④ Treatment D of winter planting Chinese milk vetch， in which the nitrogen application rate in the late rice season was 150 kg/hm 2; ⑤ Treatment E of winter planting Chinese milk vetch， in which the nitrogen application rate in the late rice season was 225 kg/hm 2. Each treatment had 3 repetitions， following the completely randomized block design. The application rates of phosphate and potassium fertilizers were the same in all the late rice treatments， in which the application rate of phosphate fertilizer was 60 kg/hm 2 and the application rate of potassium fertilizer was 75 kg/hm 2. The application rate of chemical fertilizers were calculated using urea with 46% of N， calcium magnesium phosphate with 12% of P2O5，and potassium chloride with 60% of K2O， and no treatment had the straws returned to fields after the harvest of late rice. The test plot was 11 m long and 6 m wide with an area of 66 m 2， and guarding rows were set on both sides with the width of 1 m. The test Chinese milk vetch variety was Yujiang Dayezi， which was sown on October 1， 2015 with the sowing amount of 27.5 kg/hm 2. The turning time was on March 28， 2016. Sample collection and determination
Yield measurement of Chinese milk vetch
In the flowering period of Chinese milk vetch， the 5point method was adopted to measure its fresh weight by taking 1 m 2 from each point. The total yield of fresh grass was the plot area multiplying the average yield of the 5 points.
Greenhouse gas collection and determination
The greenhouse gas was collected using static box method. The static box was 50 cm in length， width and height. The inside was made of stainless steel plate， and the outside was packed with sponge and aluminum foil insulation board of 0.5 cm thick to avoid the temperature in box rise too quickly because of the sun. There was a 12 V small fan equipped at the top of the box to prevent the gas in the box from being uneven. There was a pumping hole in the middle of the side of the box， which was controlled by a threeway valve control switch. Each plot was fixed with a sampling base， and there was a groove of 5 cm deep on the base which was used to seal with water when measuring. During the growing season of Chinese milk vetch， samples were taken every 10-15 d[19]， and the daily average temperature was recorded during the sampling period （Fig. 1）. The collection time was 08：00-11：00. Samples were collected at a time interval of 0， 10， 20， 30 min， during which a 50 mL injector was used to pump the gas inside the box for 5-10 times to fully mix the gas， and then 50 mL of gas was pumped and quickly taken back to the laboratory for analysis after being stored in a vacuum sampling bag. The concentration of N2O， CH4 and CO2 was determined by Agilent 7890 B gas chromatography. The detector for measuring CH4 and CO2 was FID with the detection temperature of 300 ℃ and column temperature of 60 ℃， and the carrier gas was 99.99% highpurity nitrogen gas with a flow rate of 30 mL/min. The detector for measuring N2O was ECD with the detection temperature of 300 ℃ and column temperature of 60 ℃， and the carrier gas was 99.99% highpurity argon/methane gas （95% argon + 5% methane） at a flow rate of 40 mL/min. Gas emission flux calculation formula was as follows：
F = ρ × h × dC/dt × 273/（ 273 + T） .
Where： F is the gas emission flux， mg/（m 2·h） or μg/（ m 2·h） ; ρ is the density of the gas under standard conditions， kg/m 3; h is the net height of the sampling box， m; dC/dt is the rate of change of the concentration of gas in the sampling tank per unit time; T is the average temperature in the sampling box during sampling， ℃; 273 is the constant of the gas equation. Global warming potential （GWP） converted the warming potential of the total seasonal emissions of various greenhouse gases into CO2 equivalents， and CH4 and N2O were 25 times and 298 times that of CO2 on the 100year scale[20]. The calculation formula was as follows：
GWP = 25（CH4） + 298（ N2O） + CO2.
Determination of soil organic carbon pool index
Before turning Chinese milk vetch into soil， soil samples were taken from each plot according to the 5point sampling method after fully mixing. Some soil was naturally dried for the determination of total soil organic carbon， and the other part of soil was stored in a refrigerator at 4 ℃ for the determination of active organic carbon. For the reference paddy field， the soil total organic carbon mass fraction was 16.67 g/kg， and the active organic carbon mass fraction was 2.11 g/kg. Soil organic carbon （SOC） was determined by potassium dichromateconcentrated sulfuric acid external heating method[21]; active organic carbon （AOC） was determined by 333 mmol/L potassium permanganate oxidation method[22]. The following were the calculation formulas for soil active organic carbon pool and carbon pool management index：
Carbon pool index = Soil organic carbon mass fraction （g/kg）/Referential cropland soil organic carbon mass fraction （g/kg）;
Steady carbon = Soil organic carbonActive organic carbon;
Carbon pool activity = Soil active organic carbon mass fraction （g/kg）/Steady carbon mass fraction （g/kg）;
Carbon pool activity index = Sample carbon pool activity/Referential soil carbon pool activity;
Carbon pool management index = Carbon pool index × Carbon pool activity index × 1 000.
Data processing
Excel 2007 and SPSS 17. 0 were used for statistical analysis of the test data， LSD was used to test the significance of difference between the sample mean values， and Origin was used for mapping.
Results and analysis
Effect of nitrogen application in late rice season on the yield of fresh Chinese milk vetch
As shown in Fig. 2， different nitrogen application rates in late rice season had significant effects on the yield of fresh Chinese milk vetch. With the increase of nitrogen application rate， the yield of Chinese milk vetch increased continuously. The yield of fresh Chinese milk vetch reached the highest in treatment E of up to 18 388.97 kg/hm 2， which was 13.94% higher than that of treatment B， and the difference between the 2 treatments reached the significant level （P<0.05）. The yield of treatment D came to the second， which was 9.98% higher than that of treatment B， and the difference between the two was significant （P<0.05）. However， there was no significant difference between treatment D and treatment C. therefore， compared with no nitrogen application in late rice season， the application of nitrogen in the late rice season could significantly increase the yield of fresh winter Chinese milk vetch. Effects of nitrogen application in late rice season on greenhouse gas emissions during the growing season of Chinese milk vetch
Nitrogen application in late rice season and winter Chinese milk vetch could affect N2O emissions from paddy fields. As shown in Fig. 3， the emissions of N2O from paddy fields were small under different treatments from after sowing Chinese milk vetch to the beginning of December 2015， and N2O emissions increased gradually after December， reaching the highest in midJanuary， 2016. During that period， the emissions of treatments C， D， E were 21.80%， 27.68% and 24.40% higher than those of treatment A， and 6.90%， 14.71%， 12.17% higher than those of treatment B. Afterwards， the N2O emission flux of each treatment gradually decreased until tilling.
Yanqin MA et al. Effects of Nitrogen Application Rate to Late Rice on Greenhouse Gas Emissions and Soil Carbon Pool During the Growing Season of Winter Chinese Milk vetch
As shown in Fig. 4， CH4 emissions were small in the early growth stage because of the low temperature and slow growth of Chinese milk vetch， which were even less than 0， presenting as the absorption of CH4 by soil. From late December， with the gradual increase of temperature and accelerating growth of Chinese milk vetch， the biomass of the aboveground part increased gradually， and the CH4 emission flux of each treated paddy field gradually increased. In late February of the following year， there was a small emission peak of CH4 in each treated rice field， when the CH4 emission flux was the highest in treatment E， reaching 0.40 mg/（m 2·h）， an increase of 110.53% from treatment A. The difference between the two was significant （P<0.05）.
As shown in Fig. 5， nitrogen application in late rice season had a great influence on CO2 emissions. With the growth of Chinese milk vetch， the CO2 emissions in each treatment increased， especially after midFebruary， the CO2 emission flux of each treatment increased rapidly. By the end of February， there was a small peak in each treatment. Compared with treatment A， the increase of CO2 emission flux was in the range of 124.36%-277.09% for treatments B， C， D， and E in this period. Then CO2 emission flux gradually decreased， reaching the maximum before turning over at the end of March.
Effects of nitrogen application in late rice season on total greenhouse gas emissions and global warming potential in the growing season of Chinese milk vetch As shown in Table 1， the total greenhouse gas emissions during the growing season of Chinese milk vetch under treatments C， D， E were significantly higher than those under treatment A of winter idling（P<0.05）. The total accumulated emissions of N2O and CO2 and global warming potential were the highest in treatment E， which showed significant differences from those of treatment B with no nitrogen application （P<0.05）， namely 11.54%， 28.19%， 26.00% higher than those of treatment B. Among the 3 treatments of C， D， E， the emissions of N2O and CO2 and global warming potential were the lowest in treatment C， which were 8.28%， 7.63%， 8.13% lower than those of treatment E （P<0.05）. Treatment showed significant differences in N2O emissions from treatment E （P<0.05）， but there was no significant difference in the total accumulated emissions of CH4 and CO2 and global warming potentials from treatment E （P>0.05）. Therefore， nitrogen application in late rice season could significantly affect the greenhouse gas emissions during the growing season of the succeeding Chinese milk vetch. Moreover， with the increase of nitrogen application rate， the emissions of N2O and CO2 during the growing season of Chinese milk vetch also gradually increased， and the global warming potential also became stronger.
Effects of winter Chinese milk vetch on soil carbon pool
As shown in Table 2， there was no significant difference in active organic carbon and soil carbon pool activity among all the treatments （P>0.05）. Among all treatments， treatment D had the highest total organic carbon content， soluble organic carbon content and carbon pool index， all of which showed significant differences from the control treatment A （P<0.05）， increasing by 21.27%， 32.26%， 11.36%， respectively. Soil active organic carbon content and carbon pool management index of treatment C were the highest， which were 12.23% and 11.99% higher than those of treatment A. The soil carbon pool management indices of treatments B， C， D， E were all significantly higher than that of treatment A with the increase ranging from 9.54% to 11.99%. However， there was no significant difference between treatments B， C， D， E （P>0.05）. Therefore， compared with winter idling treatment， winter planting Chinese milk vetch could significantly increase soil total organic carbon content， soluble organic carbon and carbon pool index， and the soil carbon pool management index was relatively higher with the nitrogen application rate of 90 and 150 kg/hm 2 in late rice season. Correlation analysis between Chinese milk vetch yield， carbon pool management index and greenhouse gas emissions
As shown in Table 3， the nitrogen application rate was significantly correlated with the N2O emissions （P<0.05）， indicating that the input of nitrogen fertilizer could increase the N2O emissions from the paddy field. The yield of Chinese milk vetch showed a significant positive correlation with N2O and CH4 emissions （P<0.05）， and extremely significant positive correlations with active organic carbon， carbon pool management index， CO2 emissions， and global warming potential （P<0.01）， indicating that with the increase of fresh Chinese milk vetch yield， the soil active organic carbon content and carbon pool management index were gradually increased， the greenhouse gas emissions also increased during the growing season of Chinese milk vetch， and the global warming potential also became stronger. The soil carbon pool management index showed extremely significant positive correlations with and CO2 emissions and global warming potential （P<0.01）， and significant positive correlations with N2O and CH4 emissions （P<0.05）. Therefore， changes in soil carbon content could affect greenhouse gas emissions in paddy fields.
Discussion
Fertilization and winter Chinese milk vetch affecting greenhouse gas emissions from paddy fields
Nitrogen fertilizer plays an important role in increasing grain yield， and it also produces a large amount of greenhouse gases such as N2O， CH4 and CO2. Studies have shown that the application of chemical nitrogen fertilizer is the main factor promoting N2O emissions from farmland， and the decisive degree of nitrate nitrogen on farmland N2O emissions is 65%[23]. The results of this study show that the total amount of N2O emissions under the treatments of winter planting Chinese milk vetch is greater than those of winter idling treatment， which is similar to the results of previous studies[17-18， 24]. Moreover， the results also show that the N2O emission flux under each treatment is higher in the late growth stage than that in the early stage， which is consistent with the results of OHara et al.[25]. The main reason is that the soil microbial activities gradually increase with the temperature， and because of the welldeveloped root system of Chinese milk vetch， the biomass of the aboveground part also increases with the increase of nitrogen application rate in the late rice season， and its physiological activity increases， resulting in the increase of N2O emissions. This view is also confirmed by a significant positive correlation between N2O emissions and the yield of fresh Chinese milk vetch. This study shows that CH4 emissions were low before midJanuary 2016， and even negative emissions appeared at certain times， while CH4 emission flux increased significantly after midDecember， which is consistent with the results of Tang et al.[17， 26]. The main reason is that CH4 emissions from paddy fields are closely related to soil water content[27] and temperature[28-29]. The temperature gradually increases after late December， and the suitable temperature， increased precipitation and good soil environment are conducive to the growth of methanogens， leading to a small peak of at least once in the CH4 emissions during the growing season of the milk vetch. The temperature drops after midJanuary， and the CH4 emissions also gradually decrease. After February， as the temperature increases， the CH4 emissions show a peak again. Winter Chinese milk vetch improving farmland soil carbon management index
The application of organic fertilizer or the combined application of organic fertilizer with inorganic fertilizer directly inputs exogenous organic matters into the soil， which can significantly increase the organic carbon content[30-31]. It has found that soil carbon pool management indices of the winter idling control are all the lowest， and the winter planting of Chinese milk vetch significantly increases the soil organic carbon and soil carbon pool management index， mainly because the notillage and direct seeding of Chinese milk vetch， on the one hand， can increase soil organic carbon storage， and on the other hand， it can increase the ground coverage， reduce soil disturbance times and soil nutrients loss， lower the erosion of wind and rain to soil， and supplement some lost carbon to the soil， which in turn enhances soil carbon sequestration capacity[33]， increases soil carbon pool activity， carbon pool activity index， and carbon stock index. At the same time， the test results also show that the content of soil active organic carbon under winter Chinese milk vetch treatments is significantly higher than that in the winter idling control， and the correlation analysis results also show that there is a significant positive correlation between the yield of fresh Chinese milk vetch and soil active organic carbon. The main reason is that the yield of fresh Chinese milk vetch gradually increases with the increase of nitrogen application rate under different nitrogen application treatments in late rice season， which improves soil active organic carbon content.
Close relations between paddy field greenhouse gas emissions and soil carbon pool
As one of the most active carbon pools in the surface ecosystem， soil carbon pool is an important release source of greenhouse gases such as N2O， CH4 and CO2， and an important sink for the greenhouse gases[34]. The increase of soil organic carbon storage in paddy fields is of great significance for decreasing soil greenhouse gas emissions， reducing atmospheric greenhouse gas concentrations， and alleviating current global warming problems. The results of this study show that there is a significant positive correlation between N2O and CH4 emissions and soil carbon pool management index （P<0.05）， and CO2 emissions and global warming potential show extremely significant positive correlations with soil carbon pool management index （P<0.01）. The reason is that notillage and direct seeding of Chinese milk vetch reduce the number of tillage in the paddy field， and the soil mineralization rate is increased during the growth of Chinese milk vetch. Moreover， the soil can be kept warm during the growth period of Chinese milk vetch. In this way， it accelerates the decay of soil organic matter， increases soil carbon pool management index， thereby increasing the N2O， CH4， CO2 emissions and greenhouse gas emission potential during the growing season of Chinese milk vetch. The global warming potential of Chinese milk vetch season shows extremely positive correlations with the yield of Chinese milk vetch and CH4 and CO2 emissions （P<0.01）， and it shows a significant positive correlation with N2O emissions （P<0.05）， which are similar to the results of Mosier et al.[35-36]. The reason is that the soil carbon sequestration capacity and carbon fixation amount are higher at high yield of winter Chinese milk vetch[32-33]， and the high carbon reserves enhance greenhouse gas emissions from farmland soils. Although the winter planting of Chinese milk vetch increases the greenhouse gas emissions of paddy fields， returning Chinese milk vetch to field can substitute some nitrogen fertilizers， which can reduce the greenhouse gas emissions from paddy fields to some extent， offsetting some of the greenhouse gas emission increase effect[37]. Conclusion
Nitrogen application in late rice season has a significant effect on the yield of the succeeding Chinese milk vetch， and the yield of Chinese milk vetch increases with the increase of nitrogen application rate. Winter planting of Chinese milk vetch can significantly increase soil organic carbon content and soil carbon pool management index. Nitrogen application in late rice season can increase the N2O， CH4， and CO2 emissions during the growing season of Chinese milk vetch， and enhance the greenhouse gas emission potential. Therefore， without reducing the yield of rice， reducing the amount of nitrogen fertilizer in the rice season can reduce the greenhouse gas emissions during the growing season of the succeeding Chinese milk vetch.
References
[1] ZHANG GL， ZHAO JN， SONG XL， et al. Effects of fertilization on soil organic carbon and carbon pool management index[J]. Plant Nutrition and Fertilizer Science， 2012， 18（ 2） ： 359-365.
[2] XU MG， YU R， SUN XF， et al. Effects of longterm fertilization on labile organic matter and carbon pool management index （CMI） of the typical soils of China[J]. Plant Nutrition and Fertilizer Science， 2006， 12（ 4） ： 459-465.
[3] WU HF， ZENG YH， PAN XH， et al. Effect of rice straw incorporation on rice yield and carbon pool management index under the application of farm mechanization[J]. Acta Agriculturae Universitatis Jiangxiensis （Natural Sciences Edition）， 2011， 33（ 5） ： 835-839， 879.
[4] CHEN SH， ZHU ZL， LIU DH， et al. Influence of straw mulching with notill on soil nutrients and carbon pool management index[J]. Plant Nutrition and Fertilizer Science， 2008， 14（4）： 806-809.
[5] WANG J， ZHU P， ZHANG N， et al. Effect of fertilization on soil active C and C pool management index of black soil[J]. Chinese Journal of Soil Science， 2003， 34（5）： 394-397.
[6] YU WT， ZHAO X， MA Q， et al. Effects of longterm fertilization on available carbon pool and carbon pool management index in an aquic brown soil[J]. Chinese Journal of Soil Science， 2008， 39（3）： 539-544.
[7] WEST TO， MARLAND G. A synthesis of carbon sequestration， carbon emission， and net carbon flux in agriculture： comparing tillage practices in the United States[J]. Agriculture Ecosystems & Environment， 2002， 91（1 /2 /3）： 217-232.
[8] TANG HM， XIAO XP， TANG WG， et al. Effects of straw recycling of winter covering crop on methane and nitrous oxide emissions in paddy field[J]. Acta Agronomica Sinica， 2011， 37（9）： 1666-1675.