Comparison of N2O Emissions From Cold Waterlogged and Normal Paddy Fields

Paddy fields are major sources of atmospheric N2O. Soil temperature and moisture strongly affect N2O emissions from rice fields. However, N2O emissions from cold-waterlogged paddy fields (CW), an important kind of paddy soil in China, are not well studied so far. It is unclear whether the N2O emissions from cold-waterlogged paddy fields are the same as normal paddy fields (NW). We investigated the N2O emission characteristics from the CW and NW paddy fields under with (R1) and without (R0) rice in Tuku Village, Baisha Town, Yangxin County (YX site, monitoring in 2013) and Huandiqiao Town, Daye City (DY site, monitoring in 2014); compared the difference and influencing factors between the CW and NW paddy fields at two sites in South China. The results showed that the N2O emissions from NWR0 were 13.4 times higher than from CWR0, and from NWR1 were 10.3 times higher than from CWR1 in the YX site. The N2O emissions from NWR0 were 2.4 times higher than from CWR0, and from NWR1 were 17.3 times higher than from CWR1 in the DY site. The structural equation models (SEMs) showed that the N2O emissions are mainly driven by rice planting and soil moisture in the NW fields at the annual scale, while soil temperature in the CW fields. Overall, N2O emissions from cold waterlogged paddy fields are significantly lower than those of normal paddy fields due to the low temperature and higher water content; however, there are dinitrogen emissions from cold waterlogged paddy fields denitrification should be further examined.


INTRODUCTION
Nitrous oxide (N 2 O) is the third-largest long-lived greenhouse gas following CO 2 and CH 4 . The lifetime of N 2 O in the atmosphere is about 121 years, and its greenhouse effect is 265 times that of CO 2 on a hundred-year scale (IPCC, 2014). Farmland ecosystems are the primary anthropogenic source of N 2 O emissions.
Rice is a staple food and feeds nearly 50% of the global population (Alexandratos and Bruinsma, 2012). Paddy fields are an important source of N 2 O emissions, and 8-11% of China's agricultural N 2 O emissions were estimated from rice fields (Zou et al., 2009). A cold-waterlogged paddy field is a major type of lowyield paddy soil in China, accounting for 15.2% of the total paddy fields in this country (Xie et al., 2015). Its main characteristics are higher groundwater levels and lower soil temperature than normal paddy fields (Qiu et al., 2013;Liu et al., 2016). Those environments make strong anaerobic conditions, poor soil structure, high organic matter contents, and low rates of N mineralization (Xie et al., 2015). Those properties of CW fields result in significantly lower rice biomass yields and higher methane emissions than normal paddy fields (NW fields) (Xu et al., 2020).
Soil water content has a decisive influence on the process of nitrification and denitrification (Davidson and Verchot, 2000). Soil water-saturated areas or flooding conditions hinder gas diffusion and form an anaerobic soil environment (Zhu et al., 2013). Alternating wet and dry, the most common water management measures in normal rice fields, causes repeated nitrification and denitrification and results in a large amount of N 2 O production and emission Hofstra and Bouwman (2005), Hu et al. (2015b), Patrick and Wyatt (1964), Fierer and Schimel (2002), Gaihre et al. (2017), Islam et al. (2018), and N 2 O emissions from lowland rice fields showed significant spatial and seasonal variations from lowland rice fields (Gaihre et al., 2017). However, due to the high groundwater level, the effects of alternating dry and wet measures in cold-waterlogged paddy fields are far inferior to normal rice fields.
As mentioned above, there are considerable differences in soil water content, soil temperature, soil organic matter content, rice yield, and methane emissions between CW fields and NW field. However, N 2 O fluxes characteristics, total N 2 O emissions, and influencing factors of cold-waterlogged paddy fields have not been explored. We hypothesized that the cold-waterlogged paddy fields have lower N 2 O emissions than normal rice fields. The impact of rice planting on nitrous oxide emissions and the significant effect of nitrous oxide emissions should differ from normal rice fields. Therefore, this study intends to systematically monitor the coldwaterlogged paddy field's N 2 O emissions characteristics on an annual scale in two representative regions and analyze the main controlling factors that affect N 2 O emissions. It's significant to understand rice fields' total greenhouse effect, accurately assessing the N 2 O emissions of China's rice field system, and reasonably formulate the emission reduction measures of this type of rice field.

Study Site and Experimental Design
The study was conducted at two sites with different climate zones in Huangshi, Hubei Province, China. One belongs to a subtropical climate zone in Tuku Village, Baisha Town, Yangxin County (YX site, 2013), and soil-derived from acid aplite. Another is Huandiqiao Town, Daye City (DY site, 2014), a northern subtropical monsoon climate zone and soil derived from carbonatite. Soil physical and chemical properties of the surface layer soil (0-20 cm) are listed in Table 1. We conducted eight treatments, including NW planted with (NWR 1 ) or without (NWR 0 ) rice and CW planted with (CWR 1 ) or without (CWR 0 ) rice in both sites. The area of each plot with rice was 100 m 2 (10 m × 10 m), and the subplot without rice was 3 m 2 (1.5 m × 2 m). Each treatment had three replicates. Urea, calcium superphosphate, and potassium chloride were applied as nitrogen, phosphorous, and potassium fertilizers, respectively (N: P 2 O 5 : K 2 O 180: 90: 120 kg hm −2 ) at both sites. Specifically, 50% nitrogen, 100% potassium, and 100% phosphorus were applied as basal fertilizer. The remaining 30% nitrogen applied at the jointing stage, and another 20% nitrogen applied ∼15 days after full heading.

Gas Collection and Analysis
N 2 O fluxes were measured using a static chamber technique, as reported previously (Xu et al., 2020). Each static chamber consisted of three parts: a bottom base, a middle chamber, and a top chamber. The chambers were wrapped with a layer of thermal insulation material. The base's four walls were drilled at 10 cm from the top with two rows of 2-cm-diameter holes to facilitate water and fertilizer flow. The base (42 cm long × 42 cm wide × 20 cm high), with a groove around the top edge, was inserted 20 cm into the soil and remained in situ except for tillage. The middle chambers with a groove around the top edge and top chambers (42 cm long × 42 cm wide × 50 cm high) covered the base (with a volume equal to the sum of middle and top chambers).
At transplanting, we transplanted four rice plants (at the same density as outside of the chamber) in the base. The gas samples are sampled every 7-10 days in the non-rice season. During the rice planting period, gases were collected for five consecutive days; thereafter, the gases were periodically collected at 7-days intervals. For each sampling, the gas within the chamber was collected four times from 8:00-10:00 a.m., using a 30-ml gas-tight syringe at 0, 5, 10, 15, and 20 min. The samples were transported to the laboratory and analyzed within 24 h. Meanwhile, soil temperature at a depth of 5 cm was recorded using an electronic digital thermometer.
The concentrations of N 2 O in gas samples were analyzed by gas chromatography (Agilent 7890A, United States) equipped with an electron capture (ECD) for N 2 O concentration analyses at 350°C, and the carrier gas was purified N 2 . We calculate the N 2 O fluxes by making a linear regression of the gas concentration.
The N 2 O fluxes was calculated using the following formula: Where F is the N 2 O flux (ug m −2 h −1 ); ρ is the N 2 O density in the standard state (kg m −3 ); V is the effective volume of the closed chamber (m 3 ), S is the base area (m 2 ); dC/dt is the change of N 2 O concentration in the sealed chamber per unit time, and T is the average temperature in the closed section.
The N 2 O cumulative gas emissions was calculated by interpolation using the following formula (Iqbal et al., 2008): where Ec is the cumulative emissions (mg m −2 ), n is the number of observations, F i (ug m −2 h −1 ), and F i+1 (ug m −2 h −1 ) are the fluxes of the i and i+1 sampling, and t i and t i+1 are the i and i+1 sampling date.

Soil Variable Measurements
Soil temperature near the base frames was measured at a depth of 5 cm in each plot and subplot using an E278 probe-type digital thermometer (Minggao Electronics Ltd., Shenzhen, China). Topsoil samples (0-20 cm) were collected randomly from five points per plot (including the plot and subplot) using a gauge auger (3-cm inner diameter) and transported immediately to the laboratory, and then homogenized and divided into two parts. One part was dried at 105°C for 24 h to determine soil water content by gravimetric. The other part was extracted with 0.5 M K 2 SO 4 solution (soil: water 1:5) for 1 h shaking and then filtrated to determine soil mineral N (NH 4 + -N and NO 3 − -N) and dissolved organic carbon (DOC). The NH 4 + -N and NO 3 − -N were analyzed using a flow-injection auto-analyzer. The DOC was measured with a TOC analyzer (Wu et al., 2017).

Statistical Analysis
N 2 O accumulation emissions are expressed as the mean ± standard deviation (SD) from three replicates. Statistical analysis was conducted using SPSS 24 (IBM SPSS, Somers, United States). The relationship between N 2 O fluxes and environmental factors was performed in R (v3.6.1) using the "basicTrendline" packages with a single environmental factor as the independent variable and N 2 O flux as the dependent variable. The model parameter is used to select the fitting function, and the p-value and R 2 value are used to determine the final regression model. Finally, SEMs were used to analyse the direct and indirect relationships between environmental factors and the N 2 O fluxes. The first step in an SEM requires establishing an a priori model based on the known effects and the relationships among the driving variables. The piecewiseSEM package (version 2.1.0) was used to analyze SEMs. We used non-significant (p > 0.05) Fisher's C values to indicate a good fit (Ochoa-Hueso et al., 2020).

Characteristic of Environmental Factors
Regardless of rice planting, the mean soil water content of CW fields was significantly higher than that of NW fields during the monitoring period (Figure 1, p < 0.01), and rice planting has no difference at both types of fields at two sites. The average concentration of DOC for the CWR 0 and CWR 1 was significantly higher than those of the NWR 0 and the NWR 1 at the DY site ( Figure 1, p < 0.01), but no difference at the YX site. The average concentration of NO 3 − -N for the CWR 0 was significantly higher than that for the NWR 0 at the DY site ( Figure 1, p < 0.01), and the average concentration of NH 4 + -N of the CWR 1 was significantly higher than that of the NWR 1 at the DY site ( Figure 1, p < 0.01). In the same site, the CW fields' mean soil temperature was lower than that of the NW fields' during the entire monitoring period, and the differences were not statistically significant (p > 0.05). However, from July 1, 2013, to September 1, 2013, the average soil temperature of the CW fields (28.45 ± 1.98°C) was significantly lower (p < 0.001) than the NW fields (29.87 ± 1.98°C) (Figure 1 A3), and from July 1, 2014, to September 1, 2014, the average soil temperature of the CW fields (29.97 ± 1.20°C) was significantly lower (p < 0.001) than the NW fields (31.52 ± 1.74°C) (Figure 1 A3).

Characteristic of N 2 O Fluxes and Cumulative Emissions
The N 2 O emissions characteristics of CW paddy fields and NW paddy fields are shown in Figure 2. The N 2 O fluxes at the YX site are between −32.93 −778.98 μg m −2 h −1 , and the DY site is between −11.82 −93.42 μg m −2 h −1 . The NW rice field of the YX site has three obvious emission peaks without rice. The other three treatments have no emission peaks. All the treatment emission peaks of the DY site are significantly lower than the YX site under the same treatment.
The annual mean N 2 O fluxes of NWR 0 treatment are 35.29 ± 16.17 μg m −2 h −1 , and 8.91 ± 3.03 μg m −2 h −1 at YX and DY sites, respectively, and of CWR 0 treatment are 4.26 ± 1.72 and 2.10 ± 1.31 μg m −2 h −1 at YX and DY sites, respectively. The mean N 2 O fluxes from CWR 0 treatment was12.1% of that of NWR 0 treatment at the YX site and was 23.6% at the DY site, respectively. The mean N 2 O fluxes of NWR 1 treatment was 12.78 ± 2.91 μg m −2 h −1 at the YX site and was 36.00 ± 26.48 μg m −2 h −1 at the DY site, Note: OM, TN, AN, TP, AP, TK, AK, and MST indicate organic matter, total nitrogen, available nitrogen, total phosphorus, available phosphorus, total potassium, available potassium, and mean soil temperature. YX and DY mean Yangxin site and Daye site. CW and NW mean cold-waterlogged paddy fields and normal paddy fields. Different lowercase letters within a single column indicate statistically significant differences at p < 0.05 between treatments. MST is mean temperature of 5 cm soil layer during rice planting.
Frontiers in Environmental Science | www.frontiersin.org July 2021 | Volume 9 | Article 660133 respectively. The mean N 2 O fluxes of CWR 1 treatment was 3.82 ± 2.07 μg m −2 h −1 at the YX site and was 0.43 ± 1.43 μg m −2 h −1 at the DY site, respectively, and mean N 2 O fluxes from CWR 1 treatment was 29.89% of that from NWR 1 treatment at the YX site and was 1.20% at DY site, respectively. The cumulative N 2 O emissions were calculated by interpolation ( Table 2). The results showed that the N 2 O cumulative emissions from the CWR 1 treatment were the lowest at both sites. The highest N 2 O cumulative emissions were observed in NWR 0 treatment at the YX site and in NWR 1 treatment at the DY site. Regardless of rice planting, N 2 O cumulative emissions of the NW fields were significantly higher than that in the CW fields (Table 2, p < 0.05) at both sites. Rice planting significantly reduced the cumulative N 2 O emissions from the NW field at the YX site but increased dramatically at the DY site. However, rice planting had no significant effect on the cumulative N 2 O emissions from CW fields at both sites ( Table 2).

Relationships between Environmental Factors and N 2 O Emissions
For the YX site, the N 2 O fluxes decrease first and then rise with the increase of the soil temperature in the NWR 0 treatment (p < 0.001, Figure 3 A 3 ). The N 2 O fluxes decrease first and then rise with the rise of the soil water content (p < 0.001, Figure 3   FIGURE 3 | A 1 , B 1 , C 1 , D 1 , and E 1 means the relationships between N 2 O fluxes and soil temperature, soil water content, DOC contents, NH 4 + -N contents, and NO 3 − -N contents in the CWR 0 , respectively. A 2 , B 2 , C 2 , D 2 , and E 1 means the relationships between N 2 O fluxes and soil temperature, soil water content, DOC contents, NH 4 + -N contents, and NO 3 − -N contents in the CWR 1 , respectively. A 3 , B 3 , C 3 , D 3 , and E 3 means the relationships between N 2 O fluxes and soil temperature, soil water content, DOC contents, NH 4 + -N contents, and NO 3 − -N contents in the NWR 0 , respectively. A 4 , B 4 , C 4 , D 4 , and E 4 means the relationships between N 2 O fluxes and soil temperature, soil water content, DOC contents, NH 4 + -N contents, and NO 3 − -N contents in the NWR 1 , respectively.
N 2 O Emission from Rice Paddy FIGURE 4 | A 1 , B 1 , C 1 , D 1 , and E 1 means the relationships between N 2 O fluxes and soil temperature, soil water content, DOC contents, NH 4 + -N contents, and NO 3 − -N contents in the CWR 0 , respectively. A 2 , B 2 , C 2 , D 2 , and E 1 means the relationships between N 2 O fluxes and soil temperature, soil water content, DOC contents, NH 4 + -N contents, and NO 3 − -N contents in the CWR 1 , respectively. A 3 , B 3 , C 3 , D 3 , and E 3 means the relationships between N 2 O fluxes and soil temperature, soil water content, DOC contents, NH 4 + -N contents, and NO 3 − -N contents in the NWR 0 , respectively. A 4 , B 4 , C 4 , D 4 , and E 4 means the relationships between N 2 O fluxes and soil temperature, soil water content, DOC contents, NH 4 + -N contents, and NO 3 − -N contents in the NWR 1 , respectively.

DISCUSSION
Our results demonstrated that the N 2 O emissions from the CW fields are significantly lower than that of the NW fields, regardless of rice planting (p < 0.05, Table 2). In this study, the soil temperature of the CW fields is significantly lower than that of the NW rice fields during the high air temperature ( Figure 6 A3 and B3). However, there is no significant difference on an annual scale. The relationship between soil temperature and N 2 O emissions is not uniform (Zhou et al., 2018;Wang et al., 2019); this difference is mainly affected by soil moisture . N 2 O emissions from soil are affected by the interaction of multiple environmental factors under natural conditions, and the relationship between temperature and water content determines whether to promote N 2 O emissions. This may be why the relationship between a single factor and N 2 O is not consistent in our study.
The N 2 O annual cumulative emissions from the NW fields are consistent with the results of Lan et al. (2020) but smaller than those reported by Huang et al. (2019), and the CW fields' N 2 O annual emissions are lower than previous studies (Huang et al., 2019;Lan et al., 2020). The possible reason is that the soil water content in Huang's research is lower than that of the NW fields and the CW fields in this study. In this study, the soil water content of the CW fields is significantly higher than that of NW fields on the annual scale ( Figure 1). Soil moisture determines the soil's redox state (Mei et al., 2011;Blagodatskaya et al., 2014). Previous research had shown that it might reduce 30-80% of N 2 O in the deep soil layer (anaerobic layer) to N 2 before being released into the atmosphere (Clough et al., 2005). The N 2 emissions from soil denitrification are considered to be a major gaseous N loss pathway, particularly in flooded paddy fields, where the strictly anaerobic environment promotes the complete reduction of nitrate or nitrite to N 2 through the intermediates of N 2 O and NO (Davidson and Verchot, 2000;Butterbach-Bahl et al., 2013). In our study, the CW rice field has been saturated for a long time and under a strictly anaerobic state (Xu et al., 2020). The strong reduction state may lead to the complete reduction of N 2 O to N 2 (Parton et al., 1996;Zhu et al., 2014). Simultaneously, the rice biomass accumulation is lower in the CW field than in the NW fields, and lower biomass accumulation means less N 2 O emissions (Xu et al., 2020). The above two points may lead to significantly lower N 2 O emissions from CW fields.
Rice planting may provide channels for N 2 O emissions, contributing more than 70% of soil N 2 O emissions during flooding but less than 20% after drainage (Yu et al., 1997;Yan et al., 2000). In this study, rice planting promoted the N 2 O accumulative emissions in the NW field at the DY site. However, the N 2 O emissions from NWR 1 were significantly lower than that of NWR 0 at the YX site, which may be related to more weeds in the treatment, and weeds (especially Monochoria vaginalis) could lead to a large amount of N 2 O production and emission. At the same time, it may also be the N 2 O emissions from NWR 0 at the YX site were significantly   Xu et al. (2020) and the higher soil pH (Table 1), due to the N 2 O emissions from low-pH soils are significantly higher than those with high-pH soils .
N 2 O emissions from paddy fields are affected by various environmental factors (Schaufler et al., 2010;Hu et al., 2015a). Pärn et al. (2018) reported that soil NO 3 − -N and soil volumetric water content together determine the geographic differentiation of global organic soil N 2 O emissions (n 58, R 2 0.72, p < 0.001), and the relationship between soil temperature and N 2 O emissions is affected by region (Pärn et al., 2018). In the present study, the structural equation model showed that the N 2 O emissions of the same type of rice fields are significantly different between the different sites. At the same time, environmental factors have no significant direct effects on N 2 O emissions. However, there are significant direct or indirect effects between soil environmental factors in each type of paddy field, confirming the cover-up effect of regional differences on environmental factors (Pärn et al., 2018).

CONCLUSION
The CW fields' annual N 2 O cumulative emissions were significantly lower than that of the NW fields under the same climatic conditions and planting systems. N 2 O emissions from the CW fields are mainly in the flooding period after transplanting, while the NW fields are primarily in the drainage period after flooding. N 2 O emissions from the CW fields are mainly affected by soil temperature; however, they are mainly affected by rice planting and soil moisture from the NW fields. The CW fields have very low N 2 O emissions and may have gaseous nitrogen emissions by denitrification. We suggest that follow-up research should study and evaluate the gaseous nitrogen emissions, and this has certain enlightenment for the governance of environmental nitrogen pollution.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Materials, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
XX and JY conceived the idea, XX, MZ, and YX conducted experiment analyzed data, XX, MZ, and MS wrote the manuscript, RH, JY, and MS. reviewed, revised and improved the manuscript.