Responses of N₂O Production and Abundances of Associated Microorganisms to Soil Profiles and Water Regime in Two Paddy Soils

Abstract

Soil moisture is one of the critical factors affecting N₂O emissions. The water regime affects the physical and chemical properties of paddy soil in different soil layers, which, in turn, affects N₂O emissions and microbial growth. However, there are few reports on the effects of different soil layers and soil moisture conditions on N₂O emission characteristics and microbial mechanisms. A 21-day microcosm experiment was performed to research the effects of soil moisture levels (60%, 100%, and 200% water holding capacity, WHC) and different soil layers (0–10, 10–20, and 20–40 cm) on N₂O emissions in hydromorphic and gleyed paddy soils. Function microbes involved in nitrification and denitrification were determined by quantitative PCR. Moreover, the abiotic variables pH, Eh, and exchangeable Fe²⁺, Fe³⁺, NH₄⁺-N, and NO₃⁻-N were also analyzed. Results showed that N₂O emissions of gleyed paddy soil were significantly higher than that of hydromorphic paddy soil, which was consistent with the result of the abundance of nitrifier and denitrifier in the two paddy soils. Soil depth, water content, and their interaction significantly affected N₂O emission (p < 0.05). Cumulative emissions of N₂O from each layer of the two paddy soils at 100% and 200% WHC were significantly higher than that under 60% WHC (p < 0.05). N₂O emissions decreased significantly with the increase of soil depth (p < 0.05), which was consistent with the change in the abundance of soil nitrifier (AOB and AOA) and denitrifier (nirK and nosZ) function genes with soil depth. The abundance of AOB, AOA, and nirK and nosZ genes decreased significantly with soil depth (p < 0.05), but did not respond significantly to the water regime. Based on the results of redundancy analysis, the contents of Fe²⁺ and Fe³⁺ were positively correlated with N₂O emissions and the abundance of AOB, AOA, and nirK and nosZ genes. These results indicate that N₂O emissions and the abundance of associated microbes are selectively affected by soil moisture and soil layers in the two paddy soils.

1. Introduction

As one of the most important greenhouse gases, Nitrous oxide (N₂O) possesses a global warming potential 298 times greater than carbon dioxide [1]. Agricultural soil is one of the dominant sources of nitrous oxide, accounting for about 65% of the total emissions of N₂O in the air [2,3]. Paddy soils represent classical agricultural soils with unique flooding and drainage regime [4,5]. It is necessary to get a deep insight into the effects and mechanism of N₂O emissions from paddy soils [6].

N₂O from paddy soils is primarily produced by microbial-mediated denitrification or nitrification [7,8]. Nitrification-associated pathways are performed by ammonia oxidizers (AOA and AOB) through oxidizing ammonium to nitrite (NO₂⁻) and emitting N₂O under aerobic conditions. Meanwhile, N₂O also can produce via the reduction of NO₂⁻ by ammonia-oxidizing bacteria (AOB). Under anaerobic conditions, N₂O is the intermediate product of denitrification where denitrifiers reduce nitrate (NO₃⁻) to dinitrogen (N₂) through NO₂⁻, nitric oxide (NO), and N₂O. Soil moisture regulates aerobic and anaerobic conditions in soils, thus influencing the relative contribution of nitrification and denitrification processes to N₂O emissions [3,9]. Generally, nitrification is the primary process of N₂O emissions in the soil with 30% < water-filled pore space (WFPS) < 70%. However, denitrification will be the dominant pathway of N₂O production under anaerobic conditions with soil moisture content higher than 80–90% WFPS [10].

The periodic flooding, drainage, tillage, and other agricultural measures of paddy fields and the fluctuation of groundwater level make paddy soils frequently in a state of dry-wet alternation [11,12,13]. As a result, different soil layers have different water contents and aeration statuses, leading to the changes of soil redox, pH, and other properties in time and space [14,15]. The changes of pH and Eh lead to the redox reaction of Fe in paddy soils and the occurrence of leaching and deposition, causing the formation of distinct layers of Fe redox and the redistribution of Fe in different soil layers [16,17]. Fe can participate in the process of nitrification and denitrification through biological or abiotic redox cycles in both anaerobic and aerobic conditions [18,19,20]. Zhu et al. [21] found in different soils that Fe content had a more significant effect on N₂O emissions than other soil indices. Nitrate-dependent ferrous oxidation (NDFO) can co-occur through biological and abiotic pathways in anaerobic environments such as wetland soils, sediments, and anaerobic microsites. Most nitrate-reducing ferrous oxidizing microbes can reduce nitrate to NO, N₂O, and N₂ [22,23,24]. Fe can also be directly involved in nitrification through Feammox; specifically, under anaerobic conditions, Fe³⁺ is reduced while NH₄⁺ is oxidized, producing NO₃⁻ and N₂ [25,26,27]. Therefore, the distribution of iron content in different soil layers may affect N₂O emissions.

In addition, environmental changes such as the changes of soil moisture, temperature, redox, and pH can affect the composition of soil microbial communities, thus, in turn, affecting the nitrification and denitrification of soil [28,29,30,31]. Soil moisture is a crucial factor affecting the diffusion of nutrients, influencing microbial community composition. In addition, soil pH has been identified as a key regulator of N₂O emissions through shaping microbial community composition [30,31]. Zhang et al. [15] found that with the deepening of soil layers, nitrogen transformation function genes (AOA amoA, AOB amoA, narG, nirK/S, and nosZ) gradually decreased, which may be because the contents and availability of NH₄⁺, NO₃⁻, DOC decreased with the deepening of soil layers [32,33,34,35]. Chen et al. [31] found that the limited denitrification in deep soils was related to the reduced abundance of denitrifying microorganisms due to the low content of nutrients such as soluble organic carbon in deep layers.

Generally, the physicochemical properties of paddy soils in different soil layers change under the influence of water status, thus affecting the abundance and activity of nitrification and denitrification microorganisms in different soil layers. So, there may be differences in N₂O emissions in different soil layers due to their different soil moisture. However, there are few reports on the effects of different soil layers and soil moisture conditions on N₂O emission characteristics and microbial mechanisms. By studying the contribution of different soil layers to the N₂O emissions, we can improve our understanding of the function of each soil layer in N₂O emissions, which will be helpful for further understanding the process of N cycling in farmland soils. In addition, the result can provide a theoretical basis to take reasonable measures for reducing N₂O emissions from farmlands. Therefore, in this study, hydromorphic and gleyed paddy soils developed under the influence of different water conditions were selected to take samples for laboratory analysis for exploring the effects of water conditions on the emissions of N₂O from the two soils of different layers. The guiding hypothesis of this study is that (1) soil moisture affected predominant N cycling processes and emissions of N₂O in both paddy soils; (2) there are significant differences in N₂O emissions from different layers of paddy soils; and (3) the responses of N₂O emissions as well as the abundance of nitrification and denitrification microorganisms to water contents and soil layers depended on different soil characteristics.

2. Materials and Methods

2.1. Location and Soils Sampling

Two types of paddy soils used for microcosm experiment were collected from two rice fields located in Xianning, Hubei Province, China (30°1′ N, 114°22′ E), characterized by a typical subtropical monsoon climate with an annual rainfall of 949.4 mm and a mean annual temperature of 16.1 °C. One type was hydromorphic paddy soil (Hydragric Anthrosols) with good irrigation and drainage conditions, obvious reductive leaching and oxidation deposition, and visible soil layer differentiation. The other was gleyed paddy soil (Gleyic Anthrosols), which was located in the paddy field of low polder area of a reservoir, with an apparent bluish-grey layer. Both soils developed from the quaternary red earth.

The 0–40 cm plough soil profile was divided into three layers, 0–10 cm, 10–20 cm, and 20–40 cm, respectively. A soil core sampler with a 5-cm diameter was used for each layer to collect five soil cores from each of three plots. The samples of each layer were mixed thoroughly, sieved through a 20-mesh then stored at 4 °C for the later experiment. The physicochemical properties of the soils are presented in

Table 1. The geochemical properties of the two paddy soils.

Soil Types and Layers		pH	SOM (g kg−1)	NH4+-N (mg kg−1)	NO3−-N (mg kg−1)	Bulk Density (g cm−3)	Total Fe (g kg−1)	HCl-Extractable Fe (g kg−1)
Hydromorphic paddy soil	0–10 cm	6.8 b (0.02)	20.8 bc (0.43)	39.5 a (0.65)	8.4 b (0.87)	1.46 b (0.04)	31.5 b (0.50)	2.2 b (0.09)
	10–20 cm	6.9 b (0.06)	18.7 c (0.78)	28.4 c (0.55)	8.1 b (0.58)	1.54 ab (0.02)	33.0 a (0.18)	3.1 a (0.15)
	20–40 cm	7.0 a (0.01)	7.3 d (1.05)	11.2 e (0.39)	3.9 c (0.32)	1.61 a (0.03)	26.4 d (0.29)	0.4 d (0.04)
Gleyed paddy soil	0–10 cm	6.1 e (0.06)	27.1 a (2.07)	34.2 b (0.97)	14.8 a (0.43)	1.15 d (0.01)	26.8 d (0.18)	1.8 c (0.13)
	10–20 cm	6.3 d (0.03)	23.2 b (2.31)	22.5 d (0.19)	9.6 b (0.37)	1.34 c (0.01)	30.7 c (0.47)	3.1 a (0.21)
	20–40 cm	6.6 c (0.09)	7.1 d (0.88)	10.0 e (1.21)	5.6 c (0.67)	1.45 b (0.06)	33.4 a (0.69)	0.10 e (0.01)

Standard errors (n = 3 replicate samples) are shown in parentheses. Different letters represent statistically significant differences between treatments at p < 0.05.

.

2.2. Microcosm Experiment

Water holding capacity (WHC) is used to characterize the maximum soil water content that farmland soil can maintain stably, and it is often used as an indicator for the upper limit of farmland irrigation and the calculation of irrigation quota. Thus, we used different percentages of WHC to simulate different soil moisture conditions. Therefore, the microcosm experiments of the samples were conducted at 25 °C under laboratory-controlled conditions under the three designed moisture content treatments (60% WHC, 100% WHC, and 200% WHC) corresponding to the soil moisture conditions of drying, wetting, and flooding.

The soil samples were pre-incubated for 7 days at 20% soil moisture content at 25 °C for activation and stabilize the microorganisms. Then, aliquots of the activated soils corresponding to 25 g dry weight soil were placed in 250 mL culture bottles. Next, (NH₄)₂SO₄ was added at a fertilization amount of 100 mg kg⁻¹ dry soil (equivalent to 225 kg N ha⁻¹ year⁻¹). All soil microcosms consisting of 108 serum bottles were incubated for 21 days at 25 °C in darkness. Each treatment of the incubation experiments was conducted in six replicates, in which three were used for soil analysis and three for gas sampling.

The 10 mL of headspace gas was sampled at days 0, 1, 3, 5, 7, 14, and 21 during the incubation by syringe for analyzing their N₂O concentration, and the emissions of N₂O were calculated according to Wang et al. [6]. At the same time, another three replicates were destructively sampled for determining soil NH₄⁺-N, NO₃⁻-N, Fe²⁺, Fe³⁺, pH, and Eh. The soil samples from the 21st day of incubation were kept at −80 °C for DNA extraction and analysis.

2.3. Physical and Chemical Analysis

N₂O concentration was determined with gas chromatography (Agilent 7890A). NO₃⁻-N and NH4⁺-N of soils were extracted with 2 mol L⁻¹ KCl solution, and NO₃⁻-N concentration was measured by ultraviolet spectrophotometry [36]. The concentration of NH4⁺-N in the soil extracts was measured by phenol hypochlorite [37]. Fe²⁺ and Fe³⁺ were extracted with 0.5 mol L⁻¹ hydrochloric acid. The extracted Fe²⁺ and Fe³⁺ was determined using the ferrozine method [38]. The soil pH was measured using a pH meter (Sartorius, Basic pH Meter PB-10, Gottingen, Germany). Eh was determined with an oxidation-reduction potentiometer.

2.4. DNA Extraction and Real-Time Quantitative PCR

The soil DNA was extracted from 0.5 g frozen soil with the FastDNA spin kit for soil (MP Biomedicals, Irvine, CA, USA) according to the manufacturer’s instructions. DNA extracts were verified by agarose gel electrophoresis and quantified using a NanoDrop^TM-Spectrophotometer (Thermo Scientific, Waltham, MA, USA).

The real-time PCR method was used to determine the functional genes of nitrification and denitrification, including AOA amoA, AOB amoA, nirK, and nosZ. The corresponding primers and amplification procedures are shown in Table S1 (See the Supplementary Materials for details). Each amplification was performed in 20-μL reactions including 10 μL of SYBR Select Master Mix (2×) Premix Ex Taq (Takara Biotech, Dalian Co., Ltd., China), 2 μL of DNA template, and 2 μL of each primer. The amplification efficiency ranged from 98% to 110%, and R² values of standard curves were >0.99 in each reaction. The amplification specificity of each gene was verified by melting curve analysis.

2.5. Statistical analysis

The soil N₂O fluxes were calculated as follows:

$$\mathrm{P}=\frac{dc}{dt}\times \frac{v}{Mv}\times \frac{Mw}{w}\times \frac{273}{T}$$

where P (μg N kg⁻¹ h⁻¹) is the soil N₂O flux, dc/dt (μL L⁻¹ h⁻¹) is the change of N₂O concentration in headspace during the period dt (h), v (L) is the headspace volume of the bottle, Mv (L) is the volume of 1 molar of gas in its standard conditions (273 K, 1.013 × 10⁶ Pa), Mw (g) is the molar mass of N₂O, w (g) is the mass of dry soil in the bottle, and T (K) is the absolute temperature.

All statistical analyses were performed using IBM SPSS Statistics 25. Statistical differences of physicochemical properties of soils were performed with one-way ANOVA. Two-way ANOVA was conducted to evaluate effects of soil layers, soil moisture and their interactions on total N₂O emissions and amoA, nirK, and nosZ gene abundance in SPSS. The effect of soil water content, soil depth, and their interactions on the abundance of nitrification and denitrification functional genes was analyzed by Two-way ANOVA. Pearson correlation analysis was used to evaluate the relationship between N₂O emissions and soil indices, such as NH₄⁺, NO₃⁻, Fe²⁺, and Fe³⁺ content. Redundancy analysis (RDA) was performed to analyze gene abundance based on soil physicochemical properties with CANOCO 5.0.

3. Results

3.1. Soil N₂O Emissions

The N₂O emission rate depicted a trend of increasing at the first stage and then decreasing with the extension of incubation time in all treatments of both soils (Figure 1). Without considering soil layers, the mean N₂O emission rate of the 100% and 200% WHC treatments was significantly higher than that of the 60% WHC treatment. In addition, the N₂O emission rate decreased with soil depth, and the highest values appeared at 0–10 cm layers for all the treatments of both soils.

Under the same water content, total N₂O emissions decreased significantly with the soil layer deepening in both paddy soils (p < 0.05) (Figure 2). For instance, the total emissions of N₂O in 0–10 cm, 10–20 cm, and 20–40 cm soil layers of 100% WHC from hydromorphic and gley paddy soils were 3296.0, 2977.3, 7.0 μg N kg⁻¹, and 9398.4, 7836.4, 6.9 μg N kg⁻¹, respectively. In each treatment, total N₂O emissions of gleyed paddy soil were significantly higher than that of hydromorphic paddy soil.

3.2. Soil Inorganic Nitrogen

NO₃⁻-N concentration increased significantly with the incubation time in 60% and 100% WHC treatments for both paddy soils (Figure 3). However, in 200% WHC treatment, the concentration of NO₃^--N showed a slightly decreasing trend at first and then increased with incubation time. In addition, the NO₃⁻-N concentration of the two paddy soils decreased significantly with the deepening of soil layers. Without considering soil layers, the average NO₃⁻-N concentration of the two soils with 60%, 100%, and 200%WHC was 27.04, 30.25, 12.69, and 26.91, 33.22, 26.91 mg N kg⁻¹, respectively.

The concentration of NH₄⁺-N of 0–10 and 10–20 cm soil layers in the 60% and 100% WHC treatments decreased gradually with incubation time (Figure 4). Different from the trend of the 60% and 100% WHC treatments, the change of NH₄⁺-N concentration in the 200% WHC treatment was firstly increased and then decreased during the initial stage (from day 0 to day 7) of incubation. The change of NH₄⁺-N was the opposite of the trend that NO₃⁻-N concentration decreased first and then increased with time under the 200% WHC.

3.3. Dynamics of Fe Concentrations

The concentrations of Fe²⁺ of 0–10 and 10–20 cm soil layers in 60% WHC treatment of the two soils firstly increased and then decreased with the increasing incubation time (Figure 5). However, in the 100% and 200% WHC treatments, the contents of Fe²⁺ in the corresponding soil layers showed a trend of rapidly decreasing and then increasing during the initial stage, and then decreased slowly until the end of incubation. Generally, soil depth significantly influenced the contents of Fe²⁺ regardless of water conditions, and the order of Fe²⁺ contents was 10–20 cm > 0–10 cm > 20–40 cm soil layers. The Fe³⁺ contents increased gradually during the initial stage and then fluctuated slightly in the 100% and 200% WHC treatments of the two soils, and the variation of Fe³⁺ contents was opposite to that of Fe²⁺ during the initial stage.

3.4. Abundance of AOA, AOB, nosZ, and nirK

The abundance of ammonia-oxidizing archaea (AOA) and ammonia-oxidizing bacteria (AOB) amoA genes in the two soils decreased with soil depth deepening (p < 0.05) (

Table 2. Summary for the two-way ANOVA on N₂O emission and AOB, AOA, and nirK and nosZ gene abundance for the two factors (soil moisture and soil depth) in the two paddy soils.

Factors	N2O	AOB amoA	AOA amoA	nirK	nosZ
Water	˂0.001	0.016	˂0.001	0.012	0.335
Soil layers	˂0.001	˂0.001	˂0.001	˂0.001	˂0.001
Water × Soil layers	˂0.001	0.297	˂0.001	0.475	˂0.565

p ˂ 0.05 means there are significant differences between treatments.

, Figure 6). The abundance of AOB and AOA amoA genes in gleyed paddy soil was much higher than in hydromorphic paddy soil (p < 0.05). The abundance of the AOB amoA gene was higher than that of the AOA amoA gene in both soils, and AOB/AOA ratio increased gradually with the increase of water contents, from 4.9 and 5.5 in 60% WHC treatment to 25.3 and 28.4 in 200% WHC treatment (Table S2).

The abundance of bacterial denitrification genes (nirK and nosZ) was also affected by soil depth, decreasing significantly with soil depth increasing (p < 0.05) (Table 2, Figure 6). The copies of nirK and nosZ genes in gleyed paddy soil were higher than those in hydromorphic paddy soil. The abundance of nirK was higher than that of nosZ in both soils. NirK/nosZ decreased significantly with soil depth increasing (Table S2).

3.5. Correlation of Environmental Factors with N₂O Emissions and the Abundance of Nitrifiers and Denitrifiers Genes

Soil depth and water content and their interaction had significant effects on N₂O emissions from the two soils (Table 2). The N₂O emissions were significantly correlated with contents of Fe²⁺, Fe³⁺, NH₄⁺-N, and NO₃⁻-N, as well as pH and Eh, and also significantly positively correlated with the abundance of AOB amoA and nirK (p < 0.05) (

Table 3. Correlation between N₂O and measured environmental factors in the two paddy soils.

	N2O Total Emission	NH4+-N	NO3--N	Fe2+	Fe3+	pH	Eh	AOB	AOA	nirK	nosZ
N2O total emission
NH4+-N	−0.563 **
NO3−-N	0.375 **	−0.290 *
Fe2+	0.516 **	−0.465 **	0.703 **
Fe3+	0.360 **	−0.446 **	0.706 **	0.937 **
pH	−0.842 **	0.424 **	−0.418 **	−0.477 **	−0.265
Eh	0.753 **	−0.278 *	0.312 *	0.409 **	0.185	−0.931 **
AOB	0.742 **	−0.419 **	0.640 **	0.535 **	0.369 **	−0.815 **	0.724 **
AOA	0.277 *	0.023	0.756 **	0.487 **	0.394 **	−0.489 **	0.453 **	0.661 **
nirK	0.491 **	−0.111	0.668 **	0.523 **	0.373 **	−0.695 **	0.614 **	0.778 **	0.888 **
nosZ	0.336 *	−0.035	0.638 **	0.650 **	0.517 **	−0.556 **	0.536 **	0.649 **	0.874 **	0.906 **

(n = 54). * p < 0.05; ** p < 0.01.

). The AOB amoA and nirK genes were significantly correlated with Fe²⁺, Fe³⁺, NH₄⁺-N, and NO₃⁻-N contents; nosZ gene was significantly correlated with Fe²⁺ and NO₃⁻-N contents (p < 0.05) (Table 3). Redundancy analysis showed that the first two axes of RDA revealed 88.9% of the variance in the abundance of AOB, AOA, and nirK and nosZ genes between the two soils. The Monte Carlo permutation test demonstrated that variation in the abundance of AOB, AOA, and nirK and nosZ genes was significantly influenced by the pH (81.9%, p = 0.002). (Figure 7,

Table 4. Redundancy analysis (RDA) through Monte Carlo permutation test for the effects of soil properties on AOA, AOB, and nirK and nosZ gene abundance.

	Explains %	Pseudo-F	p
pH	81.9	72.6	0.002
Eh	2.2	2.1	0.146
NO3−	1	0.9	0.326
Fe3+	3.2	3.6	0.078
NH4+	0.5	0.5	0.592
Fe2+	0.2	0.2	0.79

The data in bold indicated that the effect was significant.

).

4. Discussion

Soil moisture is one of the most crucial environmental factors influencing the process of soil nitrification and denitrification, thus, affecting N₂O emissions [39,40]. Generally, the increase of soil moisture content will reduce soil O₂ content, make the soil anaerobic, increase the soil denitrification potential and rate, and slow down the nitrification process [41,42]. It is generally believed that the nitrification process is the significant pathway of N₂O production under the water condition of 30–60% WFPS, while denitrification dominates the production of N₂O when the WFPS is higher than 70% [43,44]. In this study, N₂O emissions in the 100% and 200% WHC treatments were significantly higher than those under 60% WHC (Figure 1 and Figure 2), which might be due to the different sources of N₂O generation under different water contents. N₂O was produced primarily from denitrification in both soils’ 100% and 200% WHC treatments, while the N₂O was produced mainly from nitrification in the 60% WHC treatment. The N₂O emissions in the soil environment dominated by the nitrification process were far lower than that in the soil environment dominated by the denitrification process [43,45]. Under the condition of 60% WHC, the concentration of NO₃^--N in both soils continued to increase during the incubation time, while the concentration of NH₄⁺-N decreased correspondingly, proving that nitrification occurred in this water content. However, under 200% WHC, NO₃^--N concentration showed a trend of first decreasing and then increasing, and the corresponding NH₄⁺-N concentration showed the opposite trend. Those results were consistent with the research of Wang et al. [6], indicating that denitrification occurred at higher soil moisture (200% WHC) and was accompanied by the generation of N₂O. In addition, the coupled nitrification–denitrification can occur at the aerobic–anaerobic interface, and NO₂⁻ or NO₃⁻ produced by nitrification can be directly used as substrates by the denitrifying microorganisms existing in the anaerobic or hypoxic microsites, thus producing N₂O. Ma et al. [46] found that the contribution of coupled nitrification–denitrification to N₂O emissions in paddy soil of 100% and 250% WHC were 11.5% and 6.7%, respectively. Therefore, in our study, coupled nitrification–denitrification might also occur in the two soils of 100% and 200% WHC.

The N₂O emission of gleyed paddy soil was significantly higher than that of hydromorphic paddy soil, which was consistent with the copy numbers of functional genes for nitrification and denitrification in both soils, implying that the micro-ecological environment of the two soils resulted in a significantly different abundance of functional genes of nitrification and denitrification, thus leading to the different N₂O emissions. The abundances of nitrifiers and denitrifiers in the gleyed paddy and hydromorphic paddy soil responded distinctively to soil moisture and layers, suggesting N transformation processes varied with different soil types. In both soils, the abundance of nitrifiers and denitrifiers were affected by soil depth (Table 2, Figure 6), indicating that soil layers were an essential factor influencing nitrifiers and denitrifiers’ abundance. Similarly, Antonio et al. [15] showed that soil layers significantly affected the abundance of nitrifiers and denitrifiers. We speculated that surface soil could promote diffusion and transport of nutrients in soils, providing microorganisms with key substrates, such as NH₃, NO₃⁻, and soluble organic C. Soil pH can regulate N₂O production directly through affecting microflora, particularly N-transforming bacteria. For example, it has long been recognized that nitrifiers are highly sensitive to pH. Ammonia is generally thought to be the substrate for both AOA and AOB, and its concentration exponentially declines with decreasing pH due to the ionization of ammonia to ammonium. In addition, the transcription of the nosZ gene and synthesis and functionality of N₂O reductase is inhibited at acidic pH. In our study, the redundancy analysis (RDA) showed that the first two axes of RDA revealed 88.9% of the variance in the abundance of AOB, AOA, and nirK and nosZ genes between the two soils. The Monte Carlo permutation test demonstrated that the abundance of AOB, AOA, and nirK and nosZ genes was significantly influenced by the pH (81.9%, p = 0.002) (Table 4). Iron is a micronutrient essential for various critical enzymatic processes in most organisms. Extracellular iron metabolism is linked to the utilization of energy from the oxidation and reduction of iron that drives (wholly or in part) cell biosynthesis. In our study, Fe²⁺ and Fe³⁺ were positively correlated with the abundance of the four genes, indicating that Fe directly affected both nitrifying and denitrifying microorganisms in paddy soils.

Our results showed that soil layers were a key factor for N₂O emissions from the two paddy soils with different water conditions, and higher N₂O emissions were observed at surface layers. Due to long-term water management, that might be related to the differentiation of soil physicochemical properties and the nitrogen transformation function of microorganisms in different layers of paddy soils [15]. Generally, nitrification and denitrification are mainly driven by different functional microorganisms. The abundance of nitrogen transformation-related functional genes (AOB amoA, AOA amoA, nirK, and nosZ) all decreased with the increase of soil depth (Figure 6), which may be related to the decrease of soil inorganic nitrogen and water-soluble organic carbon contents with the deepening of soil layers [15,35]. We observed that both nitrifiers and denitrifiers showed the highest abundance in the 0–10 cm soil layer, consistent with the N₂O emissions, indicating that N₂O emission was related to nitrification and denitrification. This indication was supported the significant positive correlation between N₂O emission and nitrifiers and denitrifiers (AOB amoA, AOA amoA, nirK, and nosZ) (Table 3). In addition, the abundance of AOB was higher than that of the AOA amoA gene, and the ratio of AOB/AOA amoA genes increased gradually with the increase of water contents, indicating that AOB is more adapted to higher soil water conditions [47]. The ratio of nirK/nosZ genes decreased significantly with soil depth increasing, which was consistent with the trend that N₂O emissions decreased with the increase of soil depth, indicating that the structure of denitrification function microorganisms at different soil layers was different, which directly caused different N₂O emissions. These results are in accordance with previous research, which reported close relationships between physicochemical factors and soil microbial abundances [47,48].

Paddy soil is mainly affected by water fluctuation, which leads to changes in soil moisture, redox, Fe and Mn oxides, pH, etc. The changes in physicochemical properties in the soil will further affect soil microbial communities and their nitrification and denitrification functions [31]. Fe can be utilized as terminal electron acceptors by many microorganisms, thus influencing microbial community structure and activities. Our results showed that the abundance of nitrifying and denitrifying microbial genes was significantly positively correlated with iron content, indicating that iron has an essential effect on nitrifying and denitrifying microbial activity. Moreover, Fe oxides in paddy soils are prone to causing redox changes; thus, Fe can be used as an electron donor or acceptor to participate in biological and abiotic processes of nitrification and denitrification. For example, Fe²⁺ affects N₂O emissions in neutral soil through Fe oxidation coupling with the denitrification process [49,50]. Under the conditions of 100% and 200% WHC in the two soils, the Fe²⁺ contents in the corresponding soil layers showed a trend of rapid decreasing at first and then increasing during the initial stage, and decreasing slowly until the end of incubation, which indicated that the occurrence of NO₃⁻ reduction was coupled with Fe²⁺ oxidation process. Correlation analysis showed that Fe content was significantly correlated with N₂O emissions, indicating that Fe had a critical effect on N₂O emissions. Thus, in actual field management conditions, due to the different water conditions at different layers of paddy soils, the difference in N₂O emissions caused by different biochemical factors of different layers should be considered when estimating N₂O emissions.

5. Conclusions

The N₂O emissions of gleyed paddy soil are significantly higher than that of hydromorphic paddy soil, which is related to the significant difference in the abundance of nitrification and denitrification function genes caused by the different micro-ecological environments of the two paddy soils. The main reasons for the significant difference in N₂O emissions in different layers of paddy soils are the differences in physicochemical properties and the abundance of nitrogen transformation function genes in different layers. Fe plays a vital role in the distribution of nitrifying and denitrifying microorganisms and the N₂O emissions. This study provides more theoretical support for evaluating the influencing factors of N₂O emissions in agriculture practice.