Effect of Chinese Milk Vetch (Astragalus sinicus L.) and Rice Straw Incorporated in Paddy Soil on Greenhouse Gas Emission and Soil Properties

Paddy soil is an important emission source of agricultural greenhouse gases. The excessive application of chemical fertilizer to paddy soil is one of the important reasons for high greenhouse gas emissions. Emissions can be reduced through optimized agricultural management measures. The incorporation of Chinese milk vetch (CMV) and rice straw in the field to replace some of the chemical fertilizer can reduce the emissions of greenhouse gases, but the relationship between these emissions and soil properties after the incorporation of CMV and rice straw is unclear. Through the continuous determination of greenhouse gases and the physical and chemical properties of soil, it was found that the addition of CMV and straw could increase the emissions of methane (CH4) and carbon dioxide (CO2), but nitrous oxide (N2O) emissions were lower. The effect of the combined incorporating of CMV and rice straw on soil properties was more significant than CMV alone. It was also found that CH4 and CO2 emissions were positively correlated with microbial biomass carbon and nitrogen, pH, and soil catalase and β-xylosidase activities. In practice, we can reduce greenhouse gas emissions by water and fertilizer management.


Introduction
From a global perspective, the issue of climate warming is an emerging problem, and reducing non-carbon dioxide greenhouse gas emissions is a relatively fast way to reduce the impact of greenhouse gases on climate warming [1]. The agricultural sector is an important source of artificial greenhouse gas emissions, and proper farmland management can reduce agricultural greenhouse gas emissions [2,3]. Methane is the most abundant non-CO 2 greenhouse gas [1], and rice cultivation is an important source of methane emissions in agricultural production [4]. N 2 O is one of the three most important atmospheric greenhouse gases (CO 2 , CH 4 , and N 2 O). The warming potential of N 2 O is 298 times that of CO 2 , and N 2 O is also the most important ozone-depleting gas [5,6]. Although the main greenhouse gas emitted from rice fields is methane, N 2 O cannot be ignored. Intermittent irrigation to reduce CH 4 emissions will lead to increased rice field N 2 O emissions [7]. The excessive use of nitrogen fertilizer in agricultural production is the main cause of greenhouse gas emissions [1,8], so it is important to identify farmland management measures to reduce the use of chemical fertilizers.
Chinese milk vetch (CMV; Astragalus sinicus L.) is the traditional winter crop in paddy fields in southern China. CMV, used as green manure, can effectively reduce chemical fertilizer application and increase rice yield [9] while increasing the nitrogen use efficiency and soil organic matter content [10]. Long-term rotation of rice and CMV can increase soil fertility, improve soil pH, change the soil microbial community structure, increase the number of beneficial bacteria [11,12], improve the soil ecological environment, and fix nitrogen. Incorporating rice straw in the field can improve the nitrogen use efficiency, rice yield, and soil carbon balance [13], but also increase methane emissions during the rice growing season [9,14]. Studies have found that the C:N ratio in plant residues is an important factor for N 2 O emissions; the lower the C:N ratio, the higher the N 2 O emissions [15]. However, the correlation between greenhouse gas emissions and soil physical and chemical factors under mixed incorporation of CMV and rice straw in the soil remains unclear.
Based on previous research, we collected soil samples from double-cropping rice paddy soil and conducted incubation experiments to investigate the greenhouse gas emission trend of paddy soil and its correlation with soil factors under the influence of CMV and rice straw.

Treatments and Experimental Set-Up
Soil was collected from the cultivated layer (0-20 cm) of the long-term double-cropping rice field at Dengjiabu Rice Seed Farm (116.85 • N, 28.22 • E), Yingtan, Jiangxi Province. The soil was fertile silt loam, which was naturally air-dried and sieved through a 1 mm sieve. The experiment included three treatments: no addition of CMV and rice straw as a control (NSV), the addition of CMV alone (V), and the addition of CMV and straw (SV). Each treatment was replicated three times. The initial pH of the cultivated soil in the experimental field was 5.10, and the total carbon was 17.50 g kg −1 . The total nitrogen, alkaline nitrogen, available phosphorus, and available potassium contents were 1.83 g kg −1 , 218.76 mg kg −1 , 10.31 mg kg −1 , and 58.49 mg kg −1 , respectively. The soil bulk density was 1.11 g cm −3 , and the water content was 60.93%.
The water content of the CMV during the flowering period was 88%, the organic carbon content was 350.52 g kg −1 , the nitrogen content was 25.4 g kg −1 , and the C:N ratio was 13.8. Fresh CMV and rice straw were cut into 2-3 mm segments and mixed before use.
Our study was carried out by constant temperature micro-culture under dark conditions, using glass jars (500 mL, bottom diameter 8.2 cm, top diameter 5 cm), with an air-tight rubber lid. In total, 120 g of naturally air-dried soil was placed in each bottle and incubated at 60% water-filled pore space (WFPS) for 7 days under dark conditions to restart microbial activity before the application of treatments. After 7 days of pre-cultivation, we collected the gas from each jar as the initial gas sample; then, 1.8 g fresh CMV was added to all treatments except NSV, and 0.5 g of straw was added at the same time to the SV treatment. We added 20 mL of deionized water to each treatment; this was followed by thorough mixing, and an approximate 1 mm water layer was maintained, then each jars' sample was weighed separately. The soil water content was maintained under flooded conditions by weighing the jars and adding distilled water on alternate days. All treatments were randomly distributed under the same incubation conditions, and another set of the same treatments was used for soil sampling. All sample bottles were covered with tinfoil with pinholes (to maintain ventilation) to reduce excessive moisture evaporation and were incubated at 30 ± 1 • C in a controlled chamber for 52 days (until the emission of three gases tended to be stable).

Gas Sampling and Analysis
Gas samples were collected from jars after treatment initiation. The gas sampling method was adopted as described by Maucieri et al. [16]. Two gas samples, immediately after closure (T0) and after 60 min (T1), were collected from the headspace of the jars using a 20 mL syringe. Airtight three-way stop-cock syringes were used to collect the gas samples. After sampling, the bottles were covered with tinfoil. Gas samples were analyzed within 2 days by a gas chromatograph (Agilent 7890B, Santa Clara, CA, USA) equipped with a flame ionization detector (FID) for CH 4 and CO 2 analysis, and an electronic capture detector (ECD) for N 2 O analysis. During incubation, gas was collected between 8 a.m. and 10 a.m. Gas sampling was performed after added CMV and straw on days 0, 1, 2, 3, 4, 5, 6, 8, 10, 13, 17, 24, 31, 38, 45, and 52 (the frequency of gas collection according to the change of gas emission).
Greenhouse gas fluxes were calculated by the following equation [15,17,18]: where F is the emission flux rate of CO 2 (µg·g −1 ·h −1 ), CH 4 (ng·g −1 ·h −1 ), or N 2 O (ng·g −1 ·h −1 ); P is the standard atmospheric pressure (Pa); V is the head space volume (cm 3 ); ∆C is the concentration of CO 2 (ppm), CH 4 (ppb) or N 2 O (ppb) produced during the chamber was sealed; ∆t is the time between sample collections (h); R is the general gas constant; T is the absolute value of the incubation temperature (K); M is the molecular weight of CO 2 , CH 4 or N 2 O (g·mol −1 ); and m is the soil quality based on dry weight (g). The cumulative greenhouse gas (GHG) emissions were calculated as follows: where F total is the cumulative CO 2 (µg·g −1 ), CH 4 (ng·g −1 ) or N 2 O (ng·g −1 ) emissions; D i is the sampling interval, day (d); and n is the number of gas acquisition samples [19].

Soil Sampling and Analysis
Soil samples were collected to measure nitrate nitrogen and ammonium nitrogen on days 3, 9, 34, and 52. We sampled and analyzed soil parameters on day 52.
Soil pH was measured using a 1:2.5 soil-water extraction method with a glass electrode (E-301-CF; LEICI, Shanghai, China). Concentrations of nitrate and ammonium nitrogen were assessed using a flow injection analyzer (FIA automated ion analyzer, Lachat Instruments, Loveland, CO, USA). Total carbon was measured by dry combustion with a macro elemental analyzer (Vario MAX C/N; Elementar Analysensysteme, Hanau, Germany). Total nitrogen (TN) in soil was determined by the Kjeldahl method [20]. Soil microbial biomass carbon (MBC) and microbial biomass nitrogen (MBN) were determined by chloroform fumigation and K 2 SO 4 extraction, respectively [21,22]. Catalase was determined by the scat-1-y kit (Comin Biotechnology Co., Ltd, Suzhou, China) and glucosidase by the sbxys-1-y kit (Comin Biotechnology Co., Ltd, Suzhou, China). Soil easily-oxidized organic carbon (EOC) was determined by the 333 mmol L −1 potassium permanganate oxidation method [23]. R (v. 4.0.0) [24] was used for all analysis of variance. Significant effect between treatments was tested using least significant difference (LSD). Spearman correlation analysis was conducted to find out the relationship between greenhouse gases and soil factors. Related packages including agricolae [25], ggplot2 [26], PerformanceAnalytics [27], corrplot [28], and dplyr [29] were used for data analysis and plotting.

Greenhouse Gas Dynamics
During the cultivation period, the CO 2 emissions of the SV and V treatments exhibited peak values on the second day after the addition of CMV and then decreased. The emission peaks of the SV and V treatments showed no significant difference. Emissions from the NSV treatment peaked on the first day and then continued to decline. After 30 days of incubation, the CO 2 emissions of all treatments tended to be stable ( Figure 1).
In the SV and V treatments, CH 4 emission peaks appeared from the fourth and second day, with peaks of 152.29 ± 1.11 ng g −1 h −1 and 129 ± 5.04 ng g −1 h −1 , respectively. Then, the two emission values gradually decreased; after 13 days, they tended to be stable. With respect to the NSV treatment, methane emissions were consistently low during the incubation cycle; the highest value was 1.13 ng g −1 h −1 , and a negative value appeared after 13 days (Figure 1).
In contrast to the information presented above, N 2 O emissions peaked twice. The first peak values of NSV, SV, and V were 5.49 ± 0.17 ng g −1 h −1 , 7.41 ± 0.07 ng g −1 h −1 , and 11.4 ± 0.03 ng g −1 h −1 , respectively; the second emission peak appeared on the 31st day, and the time span for maintaining high emission longer than CO 2 and CH 4 , followed by a gradual decrease and then stabilization (Figure 1).
Agronomy 2020, 10, x FOR PEER REVIEW 4 of 10 high emission longer than CO2 and CH4, followed by a gradual decrease and then stabilization ( Figure 1).

Cumulative Greenhouse Gas (GHG) Emissions
The cumulative CO2 emissions of the SV treatment reached 2.47 ± 0.11 mg g −1 , which was 2.12 times the value of the NSV treatment, and the cumulative emissions of the V treatment were 1.83 times those of the NSV treatment. The cumulative CH4 emissions of the V treatment were 11.81 ± 0.16 μg g −1 ; this value was not significantly different from that of the SV treatment and was 274 times that of the NSV treatment. The cumulative N2O emissions from the V treatment were 3.68 ± 0.01 μg g −1 , and the lowest value (SV treatment) was 2.93 ± 0.03 μg g −1 (Figure 2).

Cumulative Greenhouse Gas (GHG) Emissions
The cumulative CO 2 emissions of the SV treatment reached 2.47 ± 0.11 mg g −1 , which was 2.12 times the value of the NSV treatment, and the cumulative emissions of the V treatment were 1.83 times those of the NSV treatment. The cumulative CH 4 emissions of the V treatment were 11.81 ± 0.16 µg g −1 ; this value was not significantly different from that of the SV treatment and was 274 times that of the NSV treatment. The cumulative N 2 O emissions from the V treatment were 3.68 ± 0.01 µg g −1 , and the lowest value (SV treatment) was 2.93 ± 0.03 µg g −1 (Figure 2).

The Change Trends of Soil Nitrate and Ammonium Nitrogen
From the measured data, we observed that on the third day of culture, the concentration of ammonium nitrogen was the highest, after which the concentration continued to decline. The nitrate nitrogen concentration in the SV and V treatments reached the peak value around the 35th day of culture and then decreased. There was no obvious peak value in the NSV treatment ( Figure 3).

The Change Trends of Soil Nitrate and Ammonium Nitrogen
From the measured data, we observed that on the third day of culture, the concentration of ammonium nitrogen was the highest, after which the concentration continued to decline. The nitrate nitrogen concentration in the SV and V treatments reached the peak value around the 35th day of culture and then decreased. There was no obvious peak value in the NSV treatment ( Figure 3).

Effect of CMV and Rice Straw on Soil Properties
The addition of CMV alone or in combination with straw had positive effect on total carbon (TC), easily-oxidized organic carbon (EOC), microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), pH and the activity of catalase (CAT) and β-xylosidase. The MBC and MBN contents of SV

Effect of CMV and Rice Straw on Soil Properties
The addition of CMV alone or in combination with straw had positive effect on total carbon (TC), easily-oxidized organic carbon (EOC), microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), pH and the activity of catalase (CAT) and β-xylosidase. The MBC and MBN contents of SV were 1.95 and 1.81 times that of CK respectively. On the whole, the combined application of CMV and straw was more effective as compared to the sole application of CMV (Table 1).

Correlation between Greenhouse Gas Emissions and Soil Properties
It can be seen from Figure 4 that the cumulative CO 2 emissions were positively correlated with easily-oxidized organic carbon, MBC, MBN, pH, catalase, and β-xylosidase. Among these correlations, those with MBC and MBN were 0.9 and 0.92, respectively. The cumulative CH 4 emissions had the strongest correlation with catalase, with a correlation coefficient of 0.9, followed by MBC, MBN, and β-xylosidase. N 2 O was negatively correlated with MBC and MBN, and the correlation coefficient was 0.5.

Effect of CMV and Straw on Greenhouse Gas Emissions
The V and SV treatments significantly increased CH4 emissions compared to the NSV treatment ( Figure 1). Fresh CMV has a low C:N ratio; it readily decomposes when mixed in the soil and can rapidly provide nutrients for the next crop. However, after the addition of CMV, an interaction between organic matter and soil microorganisms occurs [30]. The mineralization of organic carbon in CMV accelerates soil respiration and increases CO2 emissions [14]. At the same time, methanogens reproduce after flooding, and CH4 emissions increase [31]. The combined incorporation of straw and CMV did not significantly increase CH4 emissions, which may be related to the high C:N ratio of the

Effect of CMV and Straw on Greenhouse Gas Emissions
The V and SV treatments significantly increased CH 4 emissions compared to the NSV treatment ( Figure 1). Fresh CMV has a low C:N ratio; it readily decomposes when mixed in the soil and can rapidly provide nutrients for the next crop. However, after the addition of CMV, an interaction between organic matter and soil microorganisms occurs [30]. The mineralization of organic carbon in CMV accelerates soil respiration and increases CO 2 emissions [14]. At the same time, methanogens reproduce after flooding, and CH 4 emissions increase [31]. The combined incorporation of straw and CMV did not significantly increase CH 4 emissions, which may be related to the high C:N ratio of the straw and the slow mineralization rate of the main component (cellulose). From the perspective of the entire incubation cycle, high CH 4 emissions last for a short period of time, and this stage is a key period for controlling methane emissions. Reasonable measures should be taken to reduce CH 4 emissions, reduce soil organic carbon loss, and increase carbon sequestration.
More than two-thirds of the N 2 O emissions from the soil can be attributed to the denitrification and nitrification processes of bacteria and fungi [32]. In this experimental cycle, the first N 2 O emission peaks occurred after the CH 4 and CO 2 emissions decreased significantly. The cumulative N 2 O emissions of the SV treatment were the lowest, while those of the V treatment were the highest. Nitrous oxide emissions from the soil were reported to be positively correlated with the content of nitrogen in the soil [33], and CMV with a low C:N ratio was more conductive to N 2 O emissions when applied alone.
From the perspective of cumulative emissions, SV is the best treatment for maintaining the soil fertility and sustainable production of paddy fields. On the one hand, the use of CMV and rice straw can reduce the application of chemical fertilizers and the environmental pollution caused by nitrogen fertilizer; on the other hand, it can achieve the goal of energy conservation and emission reduction. From the changes in ammonium nitrogen and nitrate nitrogen, it can be seen that after the CMV was mixed in the soil, it rapidly released nutrients and provided nutrients for the next crop. In addition, it has nitrogen-fixing ability, so CMV represents a suitable catch crop.

The Differences of Soil Properties and Correlations between Greenhouse Gases and Soil Properties
The results showed that the addition of CMV and rice straw changed the properties of soil. The effect of the combined incorporating of CMV and rice straw on soil properties was more significant than CMV alone. In particular, there were significant positive effects on MBC and MBN ( Table 1). The addition of organic matter increased the content of soil organic carbon and nitrogen, among which MBC and MBN were the most affected [34][35][36]. The addition of organic matter to soil not only affects soil organic carbon, but also affects the activity of soil enzymes and soil pH [37].
Greenhouse gas emissions are mainly the result of soil microbial activity [38] and are closely related to soil physical and chemical properties [39]. CO 2 and CH 4 emissions were positively correlated with the pH, which is consistent with Z.P. Wang [40]. Therefore, CH 4 emissions can be reduced by appropriately controlling soil pH in the field. CH 4 emissions were significant positively correlated with CAT, maybe since the genes coding for catalase are present and important in methanogens [41]. β-xylosidase participates in the turnover of soil organic carbon and therefore has a positive correlation with CO 2 and CH 4 emissions. Some scholars believe that β-xylosidase has no relationship with CH 4 [42], which could be the reason for the more-complicated soil environment during planting and different treatment.
We used incubated soil to analyze the changes in greenhouse gases after the CMV and rice straw were mixed in the soil. Although the greenhouse gas emissions were different from the field planting conditions, the individual contributions of CMV and straw to greenhouse gas emissions could be determined. The contribution of CMV and rice straw to soil nitrogen fertilizer and the correlation between greenhouse gases and soil physical and chemical properties have been clarified. This experiment is a good supplement to field experiments and represents the theoretical basis of an effective crop residue management strategy to reduce greenhouse gas emissions in rice fields.

Conclusions
Through the continuous monitoring of greenhouse gases in a paddy soil mixed with CMV and rice straw, the emission dynamics of greenhouse gases and the nitrogen supplied by CMV and rice straw in paddy soil were determined. After the CMV and rice straw were mixed in the soil, the CH 4 emissions quickly reached a peak value, which persisted for only a short time period. The cumulative CH 4 emissions from the V treatment were not significantly different from those measured in the SV treatment and was 274 times the value measured in the NSV treatment. However, the N 2 O emissions from the SV treatment were lower than those from the control. Greenhouse gas emissions from paddy soil were positively related to the soil physical and chemical properties such as the MBC and MBN, catalase, β-xylosidase, and pH. The effect of the combined incorporating of CMV and rice straw on soil properties was more significant than CMV alone. In modern rice production, the effects of green manure and straw on greenhouse gases can be reduced by properly adjusting the physical and chemical properties of soil, for example, we can use lime to improve soil pH [43].