Nitrogen and Water Additions Affect N₂O Dynamics in Temperate Steppe by Regulating Soil Matrix and Microbial Abundance

Abstract

Elucidating the effects of nitrogen and water addition on N₂O dynamics is critical, as N₂O is a key driver of climate change (including nitrogen deposition and shifting precipitation patterns) and stratospheric ozone depletion. The temperate steppe is a notable natural source of this potent greenhouse gas. This study uses field observations and soil sampling to investigate the seasonal pattern of N₂O emissions in the temperate steppe of Inner Mongolia and the mechanism by which nitrogen and water additions, as two different types of factors, alter this seasonal pattern. It explores the regulatory roles of environmental factors, soil physicochemical properties, microbial community structure, and abundance of functional genes in influencing N₂O emissions. These results indicate that the effects of nitrogen and water addition on N₂O emission mechanisms vary throughout the growing season. Nitrogen application consistently increase N₂O emissions. In contrast, water addition suppresses N₂O emissions during the early growing season but promotes emissions during the peak and late growing seasons. In the early growing season, nitrogen addition primarily increased the dissolved organic nitrogen (DON) levels, which provided a matrix for nitrification and promoted N₂O emissions. Meanwhile, water addition increased soil moisture, enhancing the abundance of the nosZ (nitrous oxide reductase) gene while reducing nitrate nitrogen (${\mathrm{N}\mathrm{O}}_3^{-}$-N) levels, as well as AOA (ammonia-oxidizing archaea) amoA and AOB (ammonia-oxidizing bacteria) amoA gene expression, thereby lowering N₂O emissions. During the peak growing season, nitrogen’s role in adjusting pH and ammonium nitrogen (${\mathrm{N}\mathrm{H}}_4^{+}$-N), along with amplifying AOB amoA, spiked N₂O emissions. Water addition affects the balance between nitrification and denitrification by altering aerobic and anaerobic soil conditions, ultimately increasing N₂O emissions by inhibiting nosZ. As the growing season waned and precipitation decreased, temperature also became a driver of N₂O emissions. Structural equation modeling reveals that the impacts of nitrogen and water on N₂O flux variations through nitrification and denitrification are more significant during the peak growing season. This research uncovers innovative insights into how nitrogen and water additions differently impact N₂O dynamics across various stages of the growing season in the temperate steppe, providing a scientific basis for predicting and managing N₂O emissions within these ecosystems.

1. Introduction

Nitrous oxide (N₂O) is a potent and long-lived greenhouse gas that plays a critical role in climate change and stratospheric ozone depletion [1,2,3,4]. In 2020, the global average N₂O emission coefficient reached 4.3% [5]. Grasslands, which cover approximately 49% of the Earth’s land surface [6], are especially sensitive to global climate change (nitrogen deposition and precipitation pattern change) due to their location mostly in ecologically fragile zones. In China, grasslands emit an average of 76.5 ± 12.8 Gg N of N₂O annually, with 57% originating from temperate steppe soils [7]. With global warming, intensified human-induced nitrogen deposition, and shifting precipitation patterns [8,9], accurate predictions and simulation experiments on N₂O emissions have become increasingly crucial [10,11,12].

N₂O fluxes are typically regulated by climatic factors, soil matrix, and microbes [13,14,15]. Nitrogen addition directly increases the substrates for nitrification and denitrification, thereby promoting N₂O emissions [16]. Soil ${\mathrm{N}\mathrm{H}}_4^{+}$-N content is a critical factor influencing N₂O emissions [11,17]. Soil organic matter composition also significantly affects heterotrophic microbial activity. High organic carbon levels promote microbial respiration, altering the balance between nitrification and denitrification, which influences N₂O production [18]. Nitrogen addition can lead to soil acidification via processes like NH⁴⁺ nitrification and NO³⁻ leaching, which reduce N₂O reductase activity during denitrification and increase emissions [19,20]. Lower soil pH, which affects microbial biomass and community structure, can significantly reduce nitrification and denitrification rates [21,22]. Precipitation is another key factor, as it affects soil moisture, which regulates geochemical cycles and N₂O emissions [23,24]. In well-aerated dry soils, N₂O is primarily produced through nitrification [25]. However, when the soil water content reaches saturation, the resulting anaerobic conditions favor denitrification, further promoting N₂O emissions [26]. This variability in how soil moisture influences nitrification and denitrification contributes to the uncertainty in precipitation’s effect on N₂O flux [27,28]. While many studies have reported that nitrogen and water addition can directly or indirectly affect N₂O fluxes by altering regulatory factors [12,29], the results are inconsistent. The relationship between nitrogen addition and N₂O flux can be linear, exponential, or non-linear, depending on the local environmental conditions and other nutrient constraints [11,12,30,31]. Moreover, nitrogen addition and soil acidification differently impact the abundance and activity of ammonia-oxidizing functional genes [32]. In acidic soils, ammonia-oxidizing archaea (AOA) dominate, whereas ammonia-oxidizing bacteria (AOB) dominate in neutral soils [33,34]. Conversely, some research indicates that AOB is more resistant to soil acidification [35], suggesting that high AOA abundance does not necessarily dominate nitrification activity. Some research also suggests that denitrification gene abundance is not always affected by nitrogen addition [36]. Water addition shifts the source of N₂O between nitrification and denitrification processes, stimulating or suppressing N₂O emissions in temperate steppe [37,38]. However, studies on the combined effects of nitrogen deposition and precipitation on N₂O fluxes remain insufficient, despite the fact that interactions between nitrogen and water vary with precipitation. In drought years, their effects on N₂O flux may be additive, while in normal years, they can exhibit synergistic effects [39]. Some studies have also shown that rainfall can reduce the amplifying effect of nitrogen inputs on N₂O flux [40]. Finally, the seasonal distribution of rainfall plays a crucial role in N₂O emissions; however, research on how water addition affects N₂O at different stages of the growing season is still limited. Thus, additional studies are needed to investigate the seasonal interactions between nitrogen and water inputs and their effects on N₂O emissions, particularly through their influence on soil conditions and microbial processes.

Given the lack of consensus on the response of N₂O fluxes in temperate steppe to nitrogen deposition and precipitation and that only a few studies have quantified the seasonal patterns of in situ N₂O under changes in nitrogen deposition and precipitation and the factors controlling it at different stages of the growing season, this study conducted field experiments on nitrogen and water addition in a typical temperate steppe in Inner Mongolia, China. It measured the annual fluctuations in N₂O emissions and examined the impact of long-term nitrogen and water treatments during multiple sampling dates throughout the growing season on the seasonal variation patterns of N₂O. Another objective is to assess the importance of environmental factors, soil physicochemical properties, soil microbial community structure, and function as controlling factors in regulating in situ N₂O emissions under field conditions by simulating atmospheric nitrogen deposition and precipitation patterns. We hypothesize that nitrogen and water additions are differently related to N₂O emissions at different times in semi-arid grassland ecosystems, while other factors, such as microbial activity and matrix availability in the soil, may alter the effects of nitrogen and water additions throughout the seasonal variations. Our study provides a scientific basis for predicting the trend of changes in the soil matrix and microbial community in grasslands against the background of future global changes, as well as for accurately estimating soil N₂O fluxes.

2. Materials and Methods

2.1. Study Site and Experimental Design

The research station is located in Duolun County, Xilingol League, Inner Mongolia Autonomous Region, with a geographic range of 115°50′ E to 116°55′ E and 41°46′ N to 42°39′ N. The study area was set at the base of Thirteen Mile Beach, Duolun Experimental Demonstration Research Station of Restoration Ecology, Institute of Botany, Chinese Academy of Sciences. The area has a typical continental climate of temperate semi-arid to semi-humid transition, with an average annual precipitation of 385.5 mm from 1953 to 2004 [41], mainly concentrated from June to September. The soil type is primarily chestnut-calcium soil with low fertility and a loose structure, and the soil texture is dominated by sandy loam. Natural plants in Duolun County include Stipa krylovii, Agropyron cristatum, Leymus chinensis, and Artemisia frigida [42].

Monthly variations in atmospheric temperature, precipitation, soil moisture, and soil temperature in the study area for the years 2018–2020 are depicted in Figure 1. Data on precipitation and atmospheric temperature were recorded using a flux tower at the Duolun Restoration Ecology Research Station. The total rainfall in 2020 was 270.9 mm, with six consecutive months before May receiving less than 10 mm of rain per month. The majority of the rain fell between May and August, accounting for 83.54% of the annual total, with the highest quantities in July and August. The monthly mean atmospheric temperature in 2020 varied between −14.88 °C and 17.98 °C, following a single-peak pattern, with the lowest value in January and the highest in July. The average monthly soil temperature fluctuated between −8.21 °C and 19.37 °C, while soil moisture ranged from 5.56% to 15.91%. Due to the pronounced seasonal cyclicity of N₂O emissions from grasslands [43], with the highest emissions in July [44] and precipitation and nitrogen deposition peaking in summer, the time frame of this study was May–September.

The experimental sample plot consisted of 16 small plots, each measuring 4 m × 4 m, with a 1 m buffer zone separating neighboring plots. The experiment was divided into four treatment groups: CK (0 mm yr⁻¹), N4 (4 g N m⁻² yr⁻¹), W1 (56.25 mm yr⁻¹, equivalent to 15% of the rainfall), and combined nitrogen and water addition (W1N4), with four replicates each. PVC collars (19.00 cm diameter, 10.00 cm height) were randomly buried to a depth of 5 cm within each plot for N₂O flux measurements. The nitrogen and water addition experiment began in 2017, with NH₄NO₃ as the nitrogen source. Solid NH₄NO₃ was dissolved in 10 L of water (equivalent to 0.625 mm of precipitation, less than 1% of the long-term average monthly rainfall from June to August) and applied to the plots in three separate installments in early June, July, and August, avoiding rainy periods. The specific application rates are shown in Table S1.

2.2. Soil Sampling and Physicochemical Analysis

Soil samples were collected monthly from May to September 2020. In each of the four replicates of the experimental treatments, soil cores of 0–10 cm depth were collected using a 3 cm diameter soil auger. Each small plot was sampled using the five-point sampling method, after which the samples were thoroughly mixed and passed through a 2 mm mesh sieve. Soil samples from each plot were divided into two parts: one was stored at −80 °C for molecular biology analysis and enzyme activity determination, and the other was stored at −20 °C for measuring soil physicochemical properties and microbial biomass. Soil pH was measured using a pH meter (PB-10), and soil water content was determined using the dry weight method. A 10 g soil sample was extracted using 40 mL of 0.5 M K₂SO₄ solution, and the leachate was used to determine ammonium nitrogen (${\mathrm{N}\mathrm{H}}_4^{+}$-N) and nitrate nitrogen (${\mathrm{N}\mathrm{O}}_3^{-}$-N) concentrations using an AA3 continuous flow analyzer (CFA, Bran Luebbe, Hamburg, Germany). The dissolved organic carbon (DOC) content was then measured using a total organic carbon analyzer (TOC-L CPN, Shimadzu, Kyoto, Japan).

In the field, N₂O flux was measured using a portable N₂O analyzer (Model 909-0025, LGR, USA) based on the off-axis integrated cavity output spectroscopy (Off-Axis ICOS) technique, with the flux expressed in nmol N₂O m⁻² s⁻¹. This device was used in conjunction with the SF-3000 automatic multi-path long-term soil flux measurement system, ensuring approximately 1 h 19 min of observation for each plot to obtain one flux datum. Using the multiplexer system, the N₂O flux data for the 16 plots were measured sequentially through automatic chamber opening and closing, with each chamber measurement lasting approximately 200 s. The daily N₂O flux value for each plot was calculated as the average of these measurements. This study monitored the full-time period in 2020 and excluded anomalous data due to instrument malfunction. There was a long period of missing data from June 10 to 28, which was due to the solar panels experiencing a lack of power, causing the multiplexer pumps to not operate. Therefore, the instrument was returned for service to ensure the accuracy of the subsequent data and regular maintenance. After manual watering during the observation period, measurements were taken again after a one-day interval. The temperature and volumetric water content of the soil at a depth of 0–10 cm were measured with temperature and humidity probes connected to the SR-21 chamber.

2.3. DNA Extraction and Microbial Analysis

Microbial biomass carbon (MBC) and nitrogen (MBN) were analyzed through chloroform fumigation followed by extraction with 0.5 M K₂SO₄, using a total organic carbon analyzer for measurement (TOC-L CPN, Shimadzu, Kyoto, Japan). Soil enzyme activities were measured using a Synergy H1 multi-mode microplate reader. The enzymes selected for the study included N-acetyl-β-D-glucosaminidase (NAG), β-glucosidase (BG), acid phosphatase (AP), polyphenol oxidase (POX), and peroxidase (PER) to analyze the relationship between soil carbon and nitrogen cycling and enzyme activities. The 96-well plate for hydrolase was a black opaque plate (fluorescence method), and the 96-well plate for oxidase was a transparent plate (absorbed light method), and both of them were colorimetrically compared at the excitation spectrum of 360 nm and the absorption spectrum of 450 nm in the enzyme marker to obtain the quantitative analysis results, and the enzyme activity was expressed in the unit of µmol g⁻¹h⁻¹. The structure of the microbial community was determined by the phospholipid fatty acid (PLFA) biomarker method, and the composition of the phospholipid fatty acids was analyzed by gas chromatography-triple quadrupole tandem mass spectrometry (GC) to analyze the structure of the microbial community.

Quantitative PCR (Polymerase Chain Reaction) was used to determine the number of soil nitrification functional genes AOA amoA, AOB amoA, and denitrification functional genes nirK, NirS, and nosZ selected in this study. Total soil DNA was extracted using the FastDNA TM SPIN Kit (MP Biomedicals, Santa Ana, CA, USA); the concentration and purity were detected in a micro UV spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA), and the soil microbial biomass carbon index was indicated by the DNA concentration. Microbial functional gene abundance was determined using a real-time PCR detection system. Serial dilutions of 10-fold plasmids for each target gene were detected by qPCR (Real-time Quantitative PCR Detecting System) to obtain a standard curve with an R² value greater than 0.997 [45]. Negative controls, samples, and standard series were performed in triplicate on a 48-well plate. The 20 μL PCR reaction mixture contained 10 μL of premix from TaKaRa, 0.2 μL of each primer (20 mM), and 1 μL of DNA template. Bovine serum albumin was also included to provide resistance to soil inhibitors. The amplification primers and reaction conditions for the five selected nitrification and denitrification genes are shown in Table S2 [46].

2.4. Statistical Analyses

The N₂O flux data, as well as the soil physicochemical and microbial properties, were tested for normality and showed a normal distribution. Single-factor and two-factor analyses of variance (ANOVA) were performed using SPSS version 26 to examine the effects of nitrogen and water addition, as well as their interactions, on various parameters. Post-hoc multiple comparison tests were conducted using LSD (equal variances) and Dunnett-T3 (unequal variances) to assess the significance of the differences. Pearson correlation analysis and stepwise regression were used to examine the relationships between microbial abundance, soil physicochemical properties, and N₂O fluxes, as well as to identify the factors driving changes in N₂O fluxes. Principal component analysis (PCA) was applied to analyze the characteristics of the soil under nitrogen and water treatments. Finally, the Structural Equation Model (SEM) of Amos 24 was applied to investigate the mechanism of nitrogen water addition treatment on N₂O flux. Graphs were generated using the Origin 2021 software.

3. Results

3.1. Effects of Environmental Factors on N₂O Flux

As shown in Figure 2, the N₂O flux in 2020 ranged from −0.208 to 0.192 nmol m⁻² s⁻¹ over the year. During the non-growing season, fluxes showed minimal variation, ranging from −0.081 to 0.143 nmol m⁻² s⁻¹ and accounting for 20.31% of the annual total. N₂O fluxes in the control sample plots remained essentially stable from March to July after the beginning of spring, fluctuating and decreasing over the same period in the water-addition treatment sample plots, and continuing to increase after the beginning of spring until September in the nitrogen addition and nitrogen-water addition sample plots (Figure 2). It is noteworthy that N₂O fluxes generally increased in the month following the nitrogen addition treatment (Figure 2).

The distribution of precipitation during the growing season increased from the early to mid-season. High N₂O emissions occurred under high soil temperatures (ST), with N₂O primarily regulated by soil moisture (SM) at this time (Figure 3). In the early growing season, N₂O emissions increased slightly with increasing ST and decreased with decreasing ST and SM in the late growing season (Figure 3). During the non-growing season, N₂O fluxes were mainly influenced by temperature changes (Figure 3). When soil moisture was low in the early growing season, soil N₂O flux reached a trough between 8 and 13 h and peaked between 17 and 20 h (Figure 4), showing a significant negative correlation between soil temperature and N₂O fluxes. During the peak growing season, when soil moisture was the highest, the diurnal dynamics of N₂O were not affected by soil temperature, and N₂O fluxes were positive (Figure 4).

Q10 values showed significant seasonal differences across treatments (Table S3). During the early growing season, Q10 values were relatively high for all treatments, indicating increased temperature sensitivity during this period. For example, CK had a Q10 of 2.251, while nitrogen addition showed a slightly higher value of 2.291. Water addition also resulted in a high Q10 of 2.358. In contrast, during the peak growing season, Q10 values were notably lower across treatments, with N4 and W1 values decreasing to 0.419 and 0.687, respectively. During the late growing season, the combined nitrogen and water addition treatment (W1N4) exhibited the highest Q10 value of 4.135. In the non-growing season, Q10 values were generally moderate, ranging from 0.793 (W1) to 1.604 (N4), suggesting a stable but reduced influence of temperature on N₂O flux.

3.2. Impact of Nitrogen and Water Addition on N₂O Flux

During the growing season of 2020, N₂O emissions were influenced by nitrogen and water addition (p < 0.01, Table S4), but there were no interactive effects between the two. However, cumulative N₂O emissions were significantly affected by nitrogen addition, water addition, and their interactions (p < 0.01, Table S4). Specifically, water addition in June significantly reduced N₂O flux (p < 0.05, Figure 5), while nitrogen addition and combined nitrogen-water treatments in August exhibited significantly higher fluxes compared to control and water treatments (p < 0.05, Figure 5). In May, plots with water addition had significantly lower soil N₂O flux compared to control and nitrogen addition treatments (p < 0.05, Figure 5). No significant differences in N₂O fluxes were observed between the treatments in the other months. Throughout the growing season, the control, nitrogen addition, and combined nitrogen-water treatments were in an emission state, while water addition was almost in balance. In comparison, nitrogen addition increased N₂O emissions by 109.38% (p < 0.01, Table S4), whereas water addition reduced them by approximately 1.5-fold (p < 0.01, Table S4). The results showed that water addition decreased N₂O release in the early growing season, increased it during the peak and late growing seasons, and nitrogen treatments increased N₂O emissions throughout the growing season.

3.3. Effects of Nitrogen and Water Addition on Soil Physicochemical Properties

Principal component analysis (PCA) revealed that the first two components explained 63.2% of the variation in soil physicochemical properties in the temperate steppe (Figure S1). Nitrogen addition significantly affected soil pH, dissolved organic nitrogen (DON), and inorganic nitrogen content, while water addition significantly impacted SM, pH, dissolved organic carbon (DOC), and inorganic nitrogen content (

Table 1. Effects of nitrogen and water addition on soil physicochemical properties and microbial biomass.

Treatment	pH	SWC (%)	DOC (mg kg−1)	DON (mg kg−1)	${\mathbf{N}\mathbf{H}}_4^{+}$-N (mg kg−1)	${\mathbf{N}\mathbf{O}}_3^{-}$-N (mg kg−1)	MBC (mg kg−1)	MBN (mg kg−1)	ST (°C)	SM (%)
CK	7.200 ± 0.027 b	0.102 ± 0.006 a	73.118 ± 5.233 bc	16.867 ± 0.884 b	5.131 ± 0.419 a	3.371 ± 0.197 ab	467.424 ± 35.889 a	68.620 ± 3.968 ab	16.841 ± 0.799 a	8.886 ± 1.172 ab
N4	6.824 ± 0.047 c	0.094 ± 0.005 a	64.699 ± 3.754 c	20.336 ± 1.310 ab	6.466 ± 0.574 a	3.697 ± 0.223 a	407.841 ± 42.284 a	53.240 ± 5.707 b	17.178 ± 0.843 a	7.554 ± 0.774 b
W1	7.316 ± 0.045 a	0.111 ± 0.008 a	91.051 ± 7.724 a	18.985 ± 1.339 ab	5.970 ± 0.487 a	2.998 ± 0.268 b	461.755 ± 42.078 a	72.699 ± 6.883 a	16.746 ± 0.750 a	11.545 ± 1.510 a
W1N4	6.933 ± 0.033 c	0.106 ± 0.007 a	87.363 ± 6.183 ab	21.183 ± 1.378 a	6.787 ± 0.865 a	2.825 ± 0.238 b	460.438 ± 34.911 a	62.337 ± 4.582 ab	16.619 ± 0.772 a	11.432 ± 1.589 a
p value
N	<0.001	n.s.	n.s.	0.026	0.082	n.s.	n.s.	0.020	n.s.	n.s.
W	0.005	n.s.	0.001	n.s.	n.s.	0.009	n.s.	n.s.	n.s.	0.014
N * W	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.	n.s.

SWC: soil water content; DOC: soluble organic carbon; DON: soluble organic nitrogen; ${\mathrm{N}\mathrm{H}}_4^{+}$-N: ammonium nitrogen; ${\mathrm{N}\mathrm{O}}_3^{-}$-N: nitrate nitrogen; MBC: microbial biomass carbon; MBN: biomass nitrogen; ST: soil temperature; SM: soil moisture. N * W represents the nitrogen-water interaction. n.s.: not significant; different lowercase letters indicate significant differences between treatments (p < 0.05). Abbreviations apply to other charts in this Dissertation.

). Compared to the control, nitrogen addition significantly reduced soil pH (p < 0.01, Table 1), but the effect of water addition on soil pH was significantly increased (p < 0.01, Table 1). There were no significant differences in soil water content (SWC) among the treatments (Table 1). Water addition significantly increased DOC by 24.53% (p < 0.01, Table 1), and the combined nitrogen-water addition also increased DOC. Nitrogen addition significantly increased DON by 20.57% (p < 0.05, Table 1). Nitrogen addition also led to an increase in ammonium nitrogen (${\mathrm{N}\mathrm{H}}_4^{+}$-N) and nitrate nitrogen (${\mathrm{N}\mathrm{O}}_3^{-}$-N), with ${\mathrm{N}\mathrm{H}}_4^{+}$-N content being higher under nitrogen addition than under water addition, while water addition significantly reduced ${\mathrm{N}\mathrm{O}}_3^{-}$-N content (p < 0.01, Table 1). There were no interactive effects of combined nitrogen-water addition on soil physicochemical properties. Water addition significantly increased soil moisture by 29.92% (p < 0.05, Table 1), but there were no significant differences in soil moisture between the different treatments. Nitrogen and water addition did not have significant effects on soil temperature.

3.4. Effects of Nitrogen and Water Addition on Soil Microbial Factors

Soil microbial biomass carbon (MBC) showed no significant variation across treatments. Nitrogen addition significantly reduced soil microbial biomass nitrogen (MBN) (p < 0.05, Table 1), while water addition exhibited an insignificant trend toward increasing MBN (Table 1). In terms of soil hydrolytic enzyme activity across different months, β-glucosidase (BG) activity in May was higher than that in other months (p < 0.05, Figure 6c), while in August, BG activity in the water addition treatment was significantly higher than that in the control and combined nitrogen-water treatments (p < 0.05, Figure 6c). There were no significant effects of nitrogen and water addition on the activities of N-acetyl-β-D-glucosaminidase (NAG) and acid phosphatase (AP) (Table S5). Regarding soil oxidative enzyme activity, water addition had a significant effect on peroxidase (POX) activity (p < 0.01, Table S5). In June and July, water addition and combined nitrogen-water addition did not significantly increase POX activity; however, in September, POX activity in these two treatments was significantly higher than that in the nitrogen-only and control groups (p < 0.05, Figure 6a). Additionally, nitrogen addition in July significantly increased peroxidase (PER) activity by 23.83% (p < 0.05, Figure 6b), and in September, all treatments had a promoting effect on PER activity. The results indicate that nitrogen addition inhibits the activity of soil oxidative enzyme POX, while water addition mitigates the inhibition of microbial activity caused by soil acidification due to nitrogen addition (p < 0.01, Table S5).

The soil microbial community varied, as shown in Figure 7, and its content increased with the increase in the month. The total phospholipid fatty acids (PLFAs) content significantly increased by 33.10% in June for the interaction addition treatment and significantly decreased by 25.05% in July for the water addition treatment (p < 0.05, Figure 7a). Fungal PLFAs in the control group were significantly higher in September compared to May, showing a 115.62% increase (p < 0.05, Figure 7b). Water addition significantly reduced fungal PLFAs in July but showed an increasing trend in other months (p < 0.05, Figure 7b). Actinobacteria (Ac) in the control group were significantly higher in September compared to May by 110.88% (p < 0.05, Figure 7e), while arbuscular mycorrhizal fungi (AMF) in the nitrogen addition treatment were significantly lower in September compared to the control group by 42.19% (p < 0.05, Figure 7f). In contrast, water addition increased AMF content in May, June, and August but decreased it by 21.05% and 5.84% in July and September, respectively (Figure 7f). For Gram-positive bacteria (G⁺) and Gram-negative bacteria (G⁻), the control group in September was significantly higher by 89.42% and 76.66%, respectively, compared to May (p < 0.05, Figure 7c,d).

In terms of soil nitrifying microbial functional genes, water addition had a significant impact on the abundance of AOA amoA and AOB amoA genes (p < 0.01, Table S6), whereas nitrogen addition primarily influenced the abundance of AOB amoA genes (p < 0.01, Table S6). No interactive effects were observed between nitrogen and water addition on the abundance of these genes (Table S6). In the temperate steppe, nitrogen addition resulted in a significantly higher abundance of AOA amoA genes compared to water addition. Conversely, the abundance of AOB amoA genes was notably lower under water addition and combined treatments than under the control and nitrogen addition treatments (p < 0.05, Table S6). Denitrification gene abundances, however, were not significantly impacted by these treatments (Table S6).

3.5. Relationships Between N₂O and Influencing Factors

Multivariate stepwise regression analysis revealed the key factors influencing N₂O emissions at different stages of the growing season (

Table 2. Main explanatory factor models for monthly N₂O flux during the growing season.

Time	Model (n = 16)	R2	p
Early growing season (May–June)	N2O = 0.708DON − 0.427SM	0.346	0.001
Peak growing season (July–August)	N2O = −0.699pH − 0.322${\mathrm{N}\mathrm{H}}_4^{+}$-N	0.466	<0.001
Late growing season (September)	N2O = −0.615pH + 0.587ST − 0.427${\mathrm{N}\mathrm{H}}_4^{+}$-N	0.519	<0.001

). In the early growing season (May and June), SM and DON had the highest explanatory power for N₂O flux variations, accounting for 34.6% of the variance (p < 0.01, Table 2). This indicates that DON is an important predictor, even when there are no significant differences in soil moisture content. During the peak growing season (July and August), pH and ${\mathrm{N}\mathrm{H}}_4^{+}$-N explained 46.6% of the N₂O flux variation (p < 0.001; Table 2), during which N₂O was mainly influenced by substrates. In the late growing season (September), pH, ST, and ${\mathrm{N}\mathrm{H}}_4^{+}$-N together explained 51.9% of the N₂O variation (p < 0.001, Table 2), with the addition of ${\mathrm{N}\mathrm{H}}_4^{+}$-N, further indicating that the seasonal variation in N₂O emissions is primarily substrate-driven rather than directly influenced by temperature. Correlation analysis showed that DON and ${\mathrm{N}\mathrm{H}}_4^{+}$-N were significantly positively correlated with ST and SM (p < 0.01, Figure 8) and significantly negatively correlated with pH (p < 0.01, Figure 8). In the early growing season, SM was significantly positively correlated with DON and nirK (p < 0.01, Figure 8) and significantly negatively correlated with AOA amoA (p < 0.01, Figure 8), while pH showed no significant correlation with nitrification-denitrification functional gene abundances. During the peak growing season, pH exhibited a significant negative correlation with AOB amoA (p < 0.01, Figure 8) and a significant positive correlation with soil microbial biomass carbon and nitrogen (p < 0.05, Figure 8). DON, ${\mathrm{N}\mathrm{H}}_4^{+}$-N, and ST were positively associated with nirK and nosZ (p < 0.01, Figure 8), while ${\mathrm{N}\mathrm{H}}_4^{+}$-N showed a negative correlation with NirS (p < 0.05, Figure 8). Additionally, SM was negatively correlated with AOB amoA gene abundance (p < 0.05, Figure 8) but showed no significant correlation with the abundance of AOA amoA genes.

The structural equation model (SEM) revealed the key pathways influencing N₂O emissions. In the drier early growing season, nitrogen addition increased N₂O flux by elevating inorganic nitrogen (IN) and DON content. Conversely, water addition reduced N₂O emissions by negatively impacting ${\mathrm{N}\mathrm{O}}_3^{-}$-N and AOB gene expression, while positively affecting nosZ expression (Figure 9a). Additionally, nitrogen addition decreased the soil pH, which inhibited nitrification genes. These pathways accounted for approximately 51% of the variation in N₂O flux during this period (Figure 9a). During the wet peak growing season, water addition inhibited nitrification by reducing the abundance of the AOB amoA gene. However, water also stimulated N₂O emissions by promoting SM and suppressing nosZ gene expression (Figure 9b). Nitrogen and water addition had a significant effect on pH, with water addition mitigating soil acidification caused by nitrogen addition. A lower pH had a significant negative effect on AOB amoA gene abundance, while AOB amoA gene expression strongly promoted N₂O emissions (Figure 9b). The nirK and NirS genes were found to have significant negative effects on N₂O flux (Figure 9b). The SEM results suggest that direct and indirect pathways can explain about 45% of the variation in N₂O flux (Figure 9b).

4. Discussion

4.1. Seasonal Dynamics of N₂O Emissions in Temperate Steppe

The temperate steppe N₂O changes in this study showed a clear seasonal pattern, with peaks in monthly dynamic uptake occurring in spring and peaks in emissions in summer. N₂O emissions were relatively stable during the non-growing season, but fluctuations in emissions increased with increasing temperature after the beginning of spring (March), and the growing season (May to September) accounted for the majority of annual N₂O emissions, aligning with findings from previous research [47]. During the non-growing season, the soils in the study area are frozen in winter, leading to weak microbial activity and slow changes in N₂O emissions [48]. As temperatures increase during the spring thaw, soil microorganisms become active, resulting in substantial N₂O emissions over a short period [49].

The distribution of precipitation during the growing season impacts N₂O emissions. The precipitation distribution for the 2020 growing season gradually increased from the early to the middle of the season (Figure 1), delaying the peak precipitation compared to previous years [50]. During the early growing season, elevated temperatures drive microbial-mediated nitrogen cycling processes, increasing soil nitrification and denitrification rates and leading to increased N₂O emissions [51,52,53]. During the peak growing season, there is a significant positive correlation between precipitation and nitrification and denitrification rates, indicating that when temperature is not limiting, precipitation and water availability in the soil limit microbial nitrogen turnover, which varies over time [54,55]. Thus, at the highest values of soil moisture, the daily dynamics of N₂O are not affected by soil temperature and are in an emissive state (Figure 4). In contrast, during the early growing season when soil moisture is low, the daily dynamics of soil N₂O fluxes form a single valley curve between 8 and 13 h, significantly negatively correlated with soil temperature, a trend similar to findings in alfalfa grasslands in eastern Gansu using the same monitoring methods [56]. An environment of reduced O₂ due to sufficient soil moisture and a lack of plant photosynthetic intensity promotes N₂O reduction, which puts N₂O in a state of uptake [57]. As sunlight intensity increases soil temperature, enhanced transpiration promotes plant water uptake, and intensified photosynthesis provides more O₂ for microbial activity [58], gradually increasing N₂O emissions. Higher Q10 values during the early and late growing seasons suggest that N₂O emissions are more responsive to temperature changes during these periods, likely due to enhanced microbial activity under favorable moisture and substrate conditions. Conversely, lower Q10 values of N₂O during the peak growing season indicate reduced temperature sensitivity, and during this period, the variation in N₂O flux is mainly influenced by pH and ${\mathrm{N}\mathrm{H}}_4^{+}$-N.

In addition to the seasonal patterns of N₂O emissions observed in the temperate steppe, it is important to consider the potential time-lag effects between treatments and microbial responses. Studies have shown that microbial communities and soil processes may exhibit delayed responses to changes in environmental conditions [59,60]. While our study focused on the direct seasonal dynamics of N₂O emissions, the possibility of time-lag effects warrants further exploration.

4.2. Response of N₂O Fluxes During Different Periods of the Growing Season in Temperate Steppe to Nitrogen and Water Additions

Nitrogen addition significantly contributed to N₂O emissions in this study, which is similar to the results of previous studies [15,61,62]. In forest ecosystems, nitrogen deposition typically leads to increased N₂O emissions, which is consistent with the patterns observed in this study for temperate grasslands, with variations depending on soil organic matter content and microbial community dynamics. For instance, forest soils with high organic carbon content tend to exhibit stronger N₂O emission responses due to enhanced denitrification activity [63]. Integrative analyses of global greenhouse gases have also demonstrated that current nitrogen deposition significantly increases soil N₂O emissions [64]. Recent research has highlighted that the contribution of organic matter to N₂O emissions following nitrate addition is not proportional to substrate-induced soil carbon priming [31]. This finding implies that nitrate-induced N₂O emissions might not solely rely on the availability of carbon substrates, but are also influenced by the dynamics of microbial nitrogen use efficiency and denitrification pathways. Nitrogen addition initially had no significant effect on soil N₂O fluxes due to plant uptake and utilization of effective nitrogen [65]. However, as nitrogen addition increased, plant nitrogen demand was met, competition between plants and nitrifying microorganisms decreased, matrix concentrations for nitrification and denitrification increased, and microbes utilizing residual inorganic nitrogen promoted the conversion of soil nitrogen to gaseous forms, thereby enhancing soil N₂O emissions [38,66,67,68]. Thus, the impact of nitrogen addition on N₂O was not significant in the early growing season, but significantly promoted N₂O emissions during the peak of the growing season.

Studies indicate that reduced precipitation suppresses N₂O emissions, while increased precipitation promotes them [24,53], as appropriate moisture conditions directly alter soil O₂ content and thus affect microbial enzyme activity, which is consistent with the pattern of the peak growing season in this study. Adding water enhances soil moisture, reduces the rate at which oxygen diffuses from the atmosphere into the soil, facilitates the breakdown of residual organic matter, and enables the release of both organic and inorganic substances into the soil [69]. Thus, water addition may promote the release of N₂O emissions by enhancing the availability of nitrogen and carbon in the soil, which serve as substrates for denitrification [70]. However, our study found that water addition during the early growing season suppressed N₂O emissions, potentially due to excessive soil moisture reducing O₂ levels, accelerating the reduction of nitrous oxide reductase, and the limited availability of nitrogen substrates forcing denitrifying bacteria to utilize atmospheric N₂O as an alternative electron acceptor to nitrate [57], thereby reducing soil N₂O emissions. Wetland ecosystems are primarily regulated by water table fluctuations, where high water saturation promotes the complete reduction of N₂O to N₂, thereby reducing the net emissions [71]. However, water addition during the early growing season restricted nitrification, suppressing the abundance of AOA amoA and AOB amoA, thereby inhibiting N₂O emissions. This leads to different patterns of water addition effects on N₂O fluxes during different periods of the growing season. Notably, the combined addition of nitrogen and water significantly increased N₂O emissions during the peak growing season (Figure 5). This may be attributed to water addition facilitating the swift transfer of added nitrogen from the surface soil to mineral nitrogen in the 0–10 cm soil layer, thereby providing abundant substrates for both nitrification and denitrification [37,72].

The coexistence of AOA amoA and AOB amoA in our study suggests that their presence is due to complementary niche differentiation rather than direct competition. AOA amoA thrive in low-ammonia, acidic conditions, while AOB amoA are better suited to nutrient-rich, alkaline environments [73,74]. During the early growing season, microbial activity and nitrogen conversion to N₂O were slower due to lower soil temperatures, with AOA amoA playing a key role. In contrast, AOB amoA dominated during the peak growing season when higher soil temperatures and ammonia concentrations, driven by fertilization and irrigation, created more favorable conditions [75].

4.3. Mechanisms of Influence of Nitrogen and Water Additions on N₂O Fluxes Variability During Different Periods of the Growing Season

Soil N₂O emissions primarily originate from microbial-mediated nitrification, denitrification, and their coupled processes [76]. In Inner Mongolia’s semi-arid grasslands, nitrification is the main pathway for N₂O production, with nitrogen substrates and O₂ availability being the dominant regulatory factors [77]. Under increased nitrogen substrates and coupled water-nitrogen conditions, substrates promoting N₂O production and the abundance of nitrification and denitrification functional genes are the main factors influencing N₂O emissions [39,78,79].

This study found that dynamic changes in N₂O flux are significantly related to soil substrates and microbial abundance. In the early growing season, nitrogen addition increased soil DON and ${\mathrm{N}\mathrm{H}}_4^{+}$-N content, enhancing nitrification substrates and leading to an increase in nitrification functional gene abundance, especially AOB amoA (Table S6, Figure 9a), which is consistent with previous research [41,79,80]. During this period, DON-facilitated heterotrophic ammonia oxidation is a major driver of N₂O formation [81,82,83]. However, ${\mathrm{N}\mathrm{H}}_4^{+}$-N had a negative effect on N₂O emissions (Table 2), possibly due to enhanced soil respiration consuming more O₂ and creating more anaerobic conditions favorable for denitrification, thus promoting the complete reduction of N₂O to N₂. During the peak of the growing season, as the degree of conversion of ${\mathrm{N}\mathrm{H}}_4^{+}$-N to ${\mathrm{N}\mathrm{O}}_3^{-}$-N increases, the ${\mathrm{N}\mathrm{O}}_3^{-}$-N produced by the nitrification process is directly utilized by denitrifying microorganisms present in the anaerobic or hypoxic space as a matrix [84], which provides suitable reaction conditions for denitrification, and thus promotes nitrification-denitrification coupling, stimulating the production of N₂O [85]. Although ${\mathrm{N}\mathrm{O}}_3^{-}$-N significantly increased nirK gene abundance (Figure 9b), its expression was negatively correlated with N₂O emissions and positively correlated with soil moisture (Figure 8). This may be due to the environmental sensitivity of the nirK and NirS genes, resulting in their roles in denitrification being influenced by wet conditions during the growing season period [46], suggesting that denitrification is more complete under high humidity conditions, which facilitates the reduction of N₂O to N₂. The high abundance of the nosZ gene further promoted this process (Table S6), indicating that increased nirK and NirS gene expression reduced N₂O emissions. Additionally, pH also affects nitrification and denitrification processes [86]. Nitrogen addition-induced soil acidification significantly inhibits microbial communities, particularly arbuscular mycorrhizal fungi [87,88,89]. Under highly acidic conditions, fungal contributions to N₂O emissions increase [90,91]. Water addition alleviates drought to some extent, thereby promoting microbial activity and enhancing nitrogen enrichment. Consequently, MBC and MBN contents were significantly lower under nitrogen addition than under water addition (Table 1), similar to previous studies [92,93].

Soil moisture can influence changes in functional gene abundance and thus control N₂O fluxes by modulating soil O₂ concentration and ${\mathrm{N}\mathrm{O}}_3^{-}$-N content [18,94,95]. The hypoxic conditions caused by water addition are highly conducive to the denitrification pathway [96], which includes the final step of reducing N₂O to N₂ in severely saturated soils [97]. Thus, moisture conditions promote the reduction of N₂O to N₂, thereby increasing N₂O absorption [98]. Our study found that in the early growing season, water addition suppressed N₂O emissions (Figure 5). Water addition decreased the ${\mathrm{N}\mathrm{O}}_3^{-}$-N levels, inhibiting the denitrification process, and thereby reducing N₂O emissions (Figure 9). Studies have shown that AOB and AOA exhibit different contributions to the ammonia oxidation process under varying soil moisture conditions, thereby influencing N₂O emissions [73]. During the peak growing season, when water is abundant, AOB activity is typically higher, leading to increased N₂O emissions. In contrast, under relatively low soil moisture conditions in the early growing season, AOA activity tends to dominate, potentially resulting in reduced N₂O emissions. However, the anoxic environment caused by water addition stimulated the nirK and nosZ genes, making the denitrification process more complete. Research shows that the nosZ gene exhibits greater sensitivity to soil moisture fluctuations compared to nirK and nirS, correlating with increased N₂O reduction under higher soil moisture [99]. N₂O absorption depends on nosZ abundance [98]. Thus, in the peak growing season, structural equation modeling showed that water addition increased soil moisture, suppressed nosZ gene abundance, and reduced the conversion of N₂O to N₂, thereby increasing N₂O emissions (Figure 9). This is contrary to the results of the early growing season, proving that different soil moisture contents act differently on the nosZ gene and that nosZ has the potential to act as an indicator gene for soil N₂O emissions (Figure 9). Overall, nitrogen and water addition had a significant impact on the abundance of AOB amoA and AOA amoA genes, but not on denitrification functional genes (Table S6), indicating that nitrification plays a dominant role in grassland ecosystems [79]. Nitrogen addition significantly increased the abundance of ammonia-oxidizing bacteria (AOB) (Table S6), suggesting that the response of AOB abundance to nitrogen deposition is more sensitive than that of AOA [78,100].

This study found that nitrogen addition significantly increased N₂O emissions, while water addition suppressed emissions in the early growing season but promoted them during the peak growing season. These findings highlight the importance of rational management of nitrogen and water resources in temperate grassland ecosystems to reduce greenhouse gas emissions and maintain ecosystem health. Given that nitrogen addition significantly reduces soil pH and affects the structure of microbial communities [101], it is essential to explore methods to reduce nitrogen deposition, such as using low-nitrogen fertilizers or adopting improved fertilization techniques, to mitigate the negative impacts on soil acidification and microbial functionality. Since the effects of water addition on N₂O emissions vary significantly across different periods, it is recommended to prioritize water supply during drought seasons while avoiding excessive irrigation during the peak growing season to reduce greenhouse gas emissions. An integrated management strategy for nitrogen and water resources can effectively optimize the carbon and nitrogen cycles in grassland ecosystems, balancing ecosystem productivity and greenhouse gas emissions [102], thereby promoting the sustainable development of ecosystems.

Although this study provides preliminary data on the short-term effects of treatments on microbial communities and N₂O emissions, several areas warrant further investigation. Future research should focus on the following directions: (1) investigating the long-term (multi-year) effects of treatments on microbial community adaptation to better understand microbial community succession and stability over time; (2) integrating isotope techniques to better partition the sources of N₂O, particularly distinguishing between biological and non-biological sources, which would enhance our understanding of microbial contributions to N₂O emissions; and (3) examining plant-microbe interactions in mediating N₂O responses, as plant root exudates may play a crucial role in regulating microbial communities and influencing N₂O emissions.

5. Conclusions

N₂O emissions from temperate steppe in Inner Mongolia showed seasonal changes, with emissions higher in the growing season (May-September) than in the non-growing season, as influenced by temperature and moisture. During this period, nitrogen addition promoted N₂O emissions, while water addition suppressed them in the early growing season and promoted emissions in the peak period, with an interaction between the two. In the early growing season, with less precipitation, DON and SM were the main drivers; during the warm and humid peak growing season, pH and ${\mathrm{N}\mathrm{H}}_4^{+}$-N became the primary factors. As precipitation gradually decreased, important factors in the late growing season included pH, ${\mathrm{N}\mathrm{H}}_4^{+}$-N, and ST. Nitrogen and water addition affect soil N₂O flux by modifying the soil matrix and microbial abundance. Nitrogen addition provides a matrix for nitrification and denitrification, thereby stimulating N₂O emissions. In the early growing season, water addition reduces the denitrification matrix and promotes nosZ, which decreases N₂O flux. However, during the peak growing season, water addition suppresses nosZ and AOB amoA while increasing the availability of organic matter, leading to higher N₂O flux. Notably, nitrification genes are more responsive to nitrogen and water addition than denitrification genes, with AOB amoA playing a dominant role in the ammonia oxidation process. Nitrogen addition significantly reduced microbial biomass content and suppressed oxidase activity, while water addition helped alleviate microbial inhibition caused by acidification due to nitrogen addition. Overall, N₂O emissions from temperate steppe are affected differently by nitrogen and water additions and are driven by the soil matrix and the response of microbes to changes in soil conditions. Investigating these factors and their mechanisms is crucial for reducing N₂O emissions under global climate change, and provides a scientific reference for the adaptation and protection of grassland ecosystems. Future research should delve deeper into the soil-plant-microbe coupling relationships and their long-term impact on N₂O emissions, consider the abundance of functional genes in both DNA and RNA amplifications, gain a more comprehensive understanding of the pathways of N₂O flux variability and nitrogen cycling processes in ecosystems, and provide more effective management strategies for sustainable ecological development.