Effects of Different Sod-Seeding Patterns on Soil Properties, Nitrogen Cycle Genes, and N₂O Mitigation in Peach Orchards

Abstract

To clarify the role and mechanism of sod-seeding patterns in improving soil fertility and mitigating nitrous oxide (N₂O) emissions in peach orchards, we conducted a study since 2023. Taking clean tillage (CK) as the control, three sod-seeding patterns—Trifolium repens–Lolium perenne mixed sowing (TPr), T. repens single sowing (Tr), and L. perenne single sowing (Pr)—were tested to analyze soil physicochemical properties, nitrogen cycle functional genes, and N₂O emission-related genes, and to explore the driving mechanism of N₂O mitigation. Results showed that all three sod-seeding patterns significantly reduced soil pH and bulk density, increased soil electrical conductivity and mean aggregate size, and improved soil nutrient status compared with CK; TPr performed best, significantly enhancing soil enzyme activities related to carbon and nitrogen cycles. Sod-seeding patterns had no significant effect on genes involved in assimilatory nitrate reduction, denitrification, or nitrification, but significantly increased dissimilatory nitrate reduction (DNRA) and nitrogen degradation gene abundances, and reduced N₂O-producing gene (amoA + amoB, nirS + nirK) abundances. Field monitoring indicated TPr reduced N₂O emissions by 34.0%, 35.7%, and 41.0%, relative to CK, Pr, and Tr, respectively. Structural equation modeling revealed that sod-seeding reduced N₂O emissions mainly by decreasing soil NH₄⁺-N content and nirS + nirK abundance. In conclusion, sod-seeding patterns improve soil fertility and mitigate N₂O emissions in peach orchards, with TPr showing the best comprehensive benefits.

1. Introduction

China’s fruit orchard sector has expanded rapidly, with global orchard area increasing by over 65 million hectares (Mha) since 2010, and China alone now accounting for 12 Mha, which is 12% growth since 2016, making it the world’s largest fruit producer [1]. The Pinggu District in Beijing, a renowned peach-producing region, exemplifies this trend, with 14,700 hectares of peach orchards contributing 80% of Beijing’s peach revenue and 34% of the district’s agricultural output. Traditional orchard management in China relies heavily on clean tillage or herbicides for weed control, combined with excessive nitrogen (N) fertilization (550 kg N·ha⁻¹·yr⁻¹) [2], leading to soil compaction, nutrient depletion, and reduced productivity [3].

Sod-seeding (integrating perennial grasses/legumes with fruit trees) has emerged as a sustainable alternative, enhancing soil organic carbon (SOC), microbial diversity, and water retention while reducing erosion [4,5]. However, its potential to mitigate nitrous oxide (N₂O)—a potent greenhouse gas (265 × CO₂ warming potential) [6] and a critical agricultural emission source—remains underexplored in Asian temperate climates. Orchards, particularly in China, receive disproportionately high N inputs (up to 1200 N·ha⁻¹·yr⁻¹ in apple orchards) [7], yet N₂O emissions from these systems are poorly quantified [8,9]. While Mediterranean-region studies on vineyard sod-seeding show conflicting results (e.g., 50% N₂O reduction vs. no effect), no systematic investigations exist for intensive peach orchards in China. Moreover, there are still significant gaps in critical knowledge regarding the underlying mechanisms. The interactive roles of specific grass species, such as legume–grass mixtures versus monocultures, the associated shifts in soil microbial pathways that govern the nitrogen cycle (e.g., nitrification, denitrification, dissimilatory nitrate reduction), and how they interact with key environmental factors like soil moisture, pH, and ammonium content to regulate N₂O fluxes remain inadequately understood [10].

Sod-seeding improves soil structure, reduces bulk density, and increases SOC in apple and peach orchards. Legume-based systems enhance N retention via symbiotic fixation [11], while grasses reduce evaporation and runoff [12]. Additionally, N₂O production is dominated by nitrification (ammonia oxidation, amoA) [13] and denitrification (nirS/K) [14,15], with soil moisture (SM) [16] and pH as critical regulators [17,18]. Recent studies link sod-seeding to altered microbial communities, including increased AOB abundance and reduced nirK expression under legume cover. However, most studies focus on Mediterranean climates in America and Europe [19], leaving temperate orchards (e.g., peach orchards in China) understudied. Regional factors like monsoon-driven waterlogging [20] or calcareous soils may alter N₂O dynamics [21]. While legume–grass mixtures theoretically optimize N retention [22] and C sequestration [23], their combined impact on N₂O fluxes and SOC persistence remains untested. For example, legumes may elevate pH, suppressing denitrification [24], while grasses could increase SM, favoring nitrate reduction.

This study aims to compare three sod-seeding systems (ryegrass monoculture—Pr, white clover monoculture—Tr, ryegrass–clover mixture—TPr) with clean tillage (CK) in peach orchards. The objectives are (1) to measure spatiotemporal N₂O fluxes and SOC changes under different grass regimes, and (2) to investigate the microbial drivers of N₂O production pathways (nitrification, denitrification) and their interaction with soil physicochemistry using metagenomic sequencing. Our hypothesis is that the leguminous–grass mixture (TPr) will enhance soil properties and decrease N₂O emissions by providing complementary nitrogen nutrition and promoting balanced microbial activity.

2. Materials and Methods

2.1. Study Site

This study was located in Liujiadian Town, Pinggu District, Beijing, at 40°16′ N, 117°01′ E and 90–120 m above sea level. The study area belongs to a temperate continental monsoon climate, with an average annual temperature of 11.5 °C and an average annual precipitation of 680 mm [25]. The soil type in the study area is cinnamon soil, and the parent material is weathered potassium-rich pyroclastic rock; the characteristics of soil profile include: 0–20 cm is the tillage layer with the texture of sandy loam, and 20–60 cm is the sediment layer with the texture of medium loam [26]. The peach variety is “Okubo”, with a tree age of 6 years and a row spacing of 3 m × 4 m. The field management measures are to apply nitrogen fertilizer (urea 0.8 kg per plant) before germination in late March, nitrogen, phosphorus, and potassium compound fertilizer (1.2 kg per plant) in late May, and base fertilizer (decomposed sheep manure 50 kg per plant plus superphosphate 1.0 kg per plant) after fruit harvesting in late October. The irrigation method is drip irrigation, with irrigation once every 15 days from the germination stage to the fruit expansion stage, with an irrigation amount of 20 m³/667 m² each time, and no irrigation in winter (December–February of the following year). The average yield of peaches is 2500–3000 kg per 667 m², and the soluble solid content of fruit is 12–14% [27,28].

2.2. Experiment Design

Four treatments were established in the peach orchard, with the control (CK) and three sod-seeding patterns. Specifically, CK was clean tillage with weeds controlled by plowing, TPr was Trifolium repens and Lolium perenne mixed sowing, Tr was Trifolium repens monoculture, and Pr was Lolium perenne monoculture (

Table 1. Sod-seeding patterns and seed amounts.

Patterns	CK	TPr	Tr	Pr
Treatments	clean tillage	Trifolium repens and Lolium perenne mixed sowing	Trifolium repens monoculture	Lolium perenne monoculture
Seed amounts	0 kg·667 m−2	0.5 and 1 kg·667 m−2	1 kg·667 m−2	2 kg·667 m−2

). The experimental plots were arranged randomly, with 3 replicates for each treatment (Figure 1).

2.3. Measurements

2.3.1. Soil Physiochemical and Enzymic Indicators

The core ring method was used to measure bulk density (BD). Undisturbed soil cores (5 cm diameter × 5 cm height) were collected, oven-dried at 105 °C for 48 h, and weighed to calculate BD (g·cm⁻³). As for electrical conductivity (EC), soil suspensions (1:2.5 soil/deionized water, w/v) were shaken for 30 min, then measured using a conductivity meter (Orion 160, Thermo Fisher). The pH was determined in a 1:2.5 soil/deionized water suspension using a glass electrode pH meter (METTLER TOLEDO FE28). Aggregates were separated via wet sieving (2 mm, 1 mm, 0.5 mm, 0.25 mm, and <0.25 mm sieves) after dispersion with sodium hexametaphosphate. MWD (mm) was calculated as the weighted average of aggregate mass across the sieves. Soil organic carbon (SOC) was measured by potassium dichromate oxidation-colorimetry, with correction for carbonate interference. Nitrate (NO₃⁻-N) and ammonium (NH₄⁺-N) nitrogen: extracted with 2 M KCl (1:5 soil/solution, w/v), then quantified via continuous-flow colorimetry. Total nitrogen (STN) was determined by Kjeldahl digestion followed by colorimetric analysis. Soil organic nitrogen (SON) was calculated as total N minus NO₃⁻-N and NH₄⁺-N. Available phosphorus (SAP) was extracted using the Olsen method (0.5 M NaHCO₃, pH 8.5) then measured via molybdate-blue colorimetry.

Sucrase (SUC) activity was determined by sucrose hydrolysis, using 3,5-dinitrosalicylic acid (DNS) to quantify the reduced sugar. Urease (URE) activity was measured through catalyzed urea hydrolysis and the released NH₃ was trapped in boric acid and detected via indophenol blue. Catalase (CAT) activity was assessed by H₂O₂ decomposition and the residual H₂O₂ was measured through UV absorbance at 240 nm. Cellulase (CEL) activity was determined by carboxymethyl cellulose (CMC) hydrolysis, and the reduced sugars were quantified using DNS.

2.3.2. Soil N₂O Flux and Accumulative Emission

The LI-7820 high-precision N₂O/H₂O analyzer (LI-COR) combined with the 8200-01 Smart Chamber (LI-COR Biosciences, Lincoln, NE, USA) was employed to quantify soil N₂O flux ([N₂O]). The system operates on optical feedback-cavity enhanced absorption spectroscopy (OF-CEAS), enabling real-time detection of N₂O concentrations with a precision of 0.2 ppb and a response time ≤ 2 s. The Smart Chamber, equipped with a pressure-balanced ventilation system and automated mixing mechanism, minimizes pressure fluctuations and ensures stable gas sampling. Concurrently with soil N₂O flux, soil temperature and soil volumetric water content were measured using a Stevens Hydraprobe connected to the smart chamber. The measurements were conducted from 9:00 to 11:00 a.m. and 3:00 to 5:00 p.m.

The measurement protocol involved placing the Smart Chamber on soil collars (20 cm diameter, 10 cm depth) installed at experimental plots. Closed-loop gas sampling connected the chamber to the LI-7820, which continuously monitored N₂O concentrations. Data were acquired at 1 Hz and processed using SoilFluxPro^TM software (LI-COR, version 5.2.0) to calculate fluxes via linear or exponential regression, depending on concentration dynamics. Calibration was performed daily using certified N₂O standards (0–100 ppm) to correct drift and ensure accuracy.

Measurement of [N₂O] was conducted 2 to 3 times monthly, with supplementary sampling performed promptly following fertilization and intense rainfall events, and the accumulative N₂O emission was calculated according to the annual mean [N₂O].

2.3.3. Soil Metagenomic Sequencing

Soil metagenomic analysis was performed using a standardized workflow to characterize the microbial community structure and functional potential. Surface soil samples were collected from experimental plots at a depth of 0–10 cm, with 3 replicates per treatment. The samples were obtained using a sterilized auger, homogenized, and then stored at −80 °C until DNA extraction in August.

Total soil microbial DNA was extracted from 0.5 g of fresh mixed soil using the E.Z.N.A.^® Soil DNA Kit (Omega Bio-TEK, Norcross, GI, USA) according to the manufacturer’s instructions. DNA quality and concentration were assessed with a NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA) prior to metagenomic sequencing. Approximately 1 μg of high-quality DNA from each sample was used for library construction and sequenced on the Illumina HiSeq 4000 platform (Illumina Inc., San Diego, CA, USA) at Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China).

Raw sequencing reads were quality-filtered using fastp (v0.20.0; https://github.com/OpenGene/fastp, accessed on 20 September 2025) to remove low-quality bases and adapters, producing high-quality clean data for subsequent analyses. The clean reads were assembled into contigs using MEGAHIT (v1.1.2; https://github.com/voutcn/megahit, accessed on 22 September 2025), with a minimum contig length of 300 bp retained after assembly. Open reading frames (ORFs) were predicted from the contigs using Prodigal (v2.6.3; https://github.com/hyattpd/Prodigal, accessed on 22 September 2025), and ORFs with a length ≥ 100 bp were translated into amino acid sequences for downstream analysis. A non-redundant gene catalog was constructed using CD-HIT (v4.7; http://weizhongli-lab.org/cd-hit/, accessed on 23 September 2025) with thresholds of 90% sequence identity and 90% coverage. High-quality reads from each sample were mapped to the non-redundant gene catalog using SOAPaligner (v2.21; https://github.com/ShujiaHuang/SOAPaligner, accessed on 23 September 2025) to calculate gene abundance profiles. Taxonomic and functional annotations of the non-redundant gene set were obtained using DIAMOND (v2.0.13; https://github.com/bbuchfink/diamond, accessed on 23 September 2025) by performing blastp searches (E-value ≤ 1 × 10⁻⁵) against multiple databases, including NR, eggNOG, KEGG, and CAZy.

2.4. Statistical Analysis

The experimental data were presented as mean ± standard deviation (n = 3) and analyzed using SPSS 25 software (IBM, Armonk, NY, USA). Tukey’s t-test was used to assess the significant differences, and p < 0.05 was considered to be statistically significant. The structural equation model was constructed using the maximum likelihood evaluation program implemented in IBM SPSS AMOS 25 software (IBM, USA).

3. Results

3.1. Effects on Soil Properties

All sod-seeding patterns were found to have significant effects on soil physicochemical properties (Bulk density, EC, pH, and MWD), nutrient contents (SOC, NO₃⁻-N, NH₄⁺-N, SON, STN, and SAP) and soil enzymes (SUC, URE, CAT, and CEL) (Figure 2).

Soil bulk density significantly decreased in all sod-seeding treatments compared to the CK, with most changes observed in TPr pattern (Figure 2a). Soil electrical conductivity (EC) elevated dramatically in all sod-seeding treatments compared to the CK, with significant difference among TPr, Tr, and Pr patterns (Figure 2b). Compared with CK, the decline of pH was significant for all sod-seeding patterns, approximating to chemical neutrality (pH = 7) (Figure 2c). The mean weight diameter (MWD) of soil aggregates in TPr, Tr, and Pr treatments was notably higher than that in CK, reflecting improved soil aggregate stability under sod-seeding patterns (Figure 2d). Generally, sod-seeding patterns modified soil physicochemical characteristics by reducing bulk density and pH and increasing EC and MWD.

Sod-seeding patterns (TPr, Tr, Pr) improved soil nutrient status. Compared to the CK, the SOC content consistently increased in all sod-seeding treatments, with significant changes observed only in the treatment of Pr (Figure 2e). The NO₃⁻-N contents were markedly higher in TPr and Pr than those in Tr and CK, while the NH⁴⁺-N content was greatest in Pr, followed by CK, Tr, and TPr, with a significant difference only between Pr and TPr (Figure 2f,g). Relative to the CK, the SON and STN were significantly boosted in both TPr and Tr treatments; however, the SON decreased but the STN increased in Pr treatment (Figure 2h,i). With significant difference, the SAP was optimal in the Tr, followed by Pr, and higher than that in TPr and CK (Figure 2j). Collectively, sod-seeding patterns enhanced soil organic and inorganic nutrients, with Pr for SOC and NH⁴⁺-N, TPr, and Tr for SON and STN, and Tr for SAP.

Sod-seeding patterns (TPr, Tr, Pr) enhanced soil enzyme activities as well. Specifically, the activities of SUC, URE, and CEL in all sod-seeding patterns were higher than those in CK, while only the increase in the TPr was significant (Figure 2k,l,n). CAT activity boosted dramatically in all sod-seeding treatments compared to the CK, with no significant difference among TPr, Tr, and Pr patterns (Figure 2m). In general, TPr pattern performed best for improvement of soil enzyme activity.

3.2. Effects on N₂O Emission

The temporal variations in soil N₂O flux ([N₂O]) showed distinct modes across sod-seeding patterns (Figure 3a). As for the relative magnitude of [N₂O] in the four treatments, the whole year can be divided into three stages. In the first stage, from January to March, which was the period of soil freezing, thawing, and grass dormancy, the [N₂O] of CK was higher than that of the three sod-seeding patterns, and the [N₂O] showed little difference among Pr, TPr, and Tr. From April to mid-June was the second stage, overlapping with grass reviving, peach tree flowering, and fruit setting. There was no obvious difference in [N₂O] among the four treatments, and the strong effects of fertilization, irrigation, and flower and fruit thinning on [N₂O] masked the influence from sod-seeding patterns. The third stage lasted from late June to December, in parallel with the sod-grass growth periods from florescence to senescence, when the [N₂O] of TPr was the lowest, followed by CK, while Pr and Tr showed similarly higher [N₂O].

From the perspective of the annually changing trend of soil nitrous oxide flux ([N₂O]), the CK was similar to TPr, showing a relatively stable yet moderate flux trajectory with reduced peak intensities throughout the year. In contrast, Pr and Tr treatments presented more pronounced peaks, exhibiting a multimodal form (Figure 3a).

As for the cumulative N₂O emission, there was no significant difference among CK, Pr, and Tr treatments, while TPr exhibited markedly lower values (Figure 3b). The cumulative N₂O emission in CK (0.39 g N₂O·m⁻²·yr⁻¹) was mainly contributed by the soil freezing and thawing process during the first stage (Figure 3a,b). Compared to CK, the insignificantly higher cumulative N₂O emissions in Pr (0.40 g N₂O·m⁻²·yr⁻¹) and Tr (0.43 g N₂O·m⁻²·yr⁻¹) were mostly determined by the third stage, relating to the active physiological activities of sod-grass (Figure 3a,b). However, when T. repens and L. perenne were combined as the TPr pattern, the cumulative N₂O emission (0.26 g N₂O·m⁻²·yr⁻¹) significantly declined by 34.0%, 35.7%, and 41.0% relative to CK, Pr, and Tr (Figure 3b). These results indicated that the TPr sod-seeding pattern can effectively reduce soil N₂O cumulative emissions. Such differences might be attributed to altered soil microbial communities, especially those related to nitrogen cycling processes under the TPr pattern.

3.3. Effects on Nitrogen Cycle Genes

Sod-seeding patterns (TPr, Tr, Pr) significantly influenced soil nitrogen (N) cycling functional genes. The abundance of genes involved in assimilatory nitrate reduction (ANRA), denitrification, and nitrification processes in the soil nitrogen cycle did not differ significantly among different sod-seeding patterns (Figure 4a,b,d). Compared with the CK, Tr, and TPr (TPr-P and TPr-T), treatments significantly increased the gene abundance of the dissimilatory nitrate reduction (DNRA) process (Figure 4c). Tr, Pr, and TPr-P significantly boosted gene abundance for the nitrogen degradation process compared to the CK, and the TPr-P treatment exhibited the highest gene abundance (Figure 4e). Both Tr and TPr (TPr-P and TPr-T) treatments promoted gene abundance for the nitrogen fixation process; however, only Tr exhibited a significant increase compared to other treatments (Figure 4f). Collectively, sod-seeding patterns, especially TPr, differentially regulated soil N cycling genes across processes, with pronounced effects on ANRA, DNRA, N degradation, and N fixation, while denitrification and nitrification were less responsive.

The sod-seeding pattern did not significantly affect the gene abundance of nitrification and denitrification processes, which in turn played a decisive role in soil N₂O emission. As a result, the response of specific gene abundance closely related to N₂O for sod-seeding patterns was further analyzed. The sod-seeding pattern had no significant effect on the abundance of nitrous oxide reductase gene (nosZ) that determines N₂O consumption, compared with the CK (Figure 5c). Nevertheless, the grass-growing pattern dramatically reduced the abundance of the ammonia monooxygenase gene (amoA + amoB) and nitrite reductase gene (nirS + nirK), which determine N₂O production, and the TPr treatment had the most prominent reduction effect (Figure 5a,b). In summary, the sod-seeding pattern differentially regulates N₂O-related gene pools: it prominently declined genes in nitrification and early denitrification steps (amoA/amoB, nirS/nirK) but made subtle impact on the terminal nosZ gene.

3.4. The Impact of Sod-Seeding Pattern on N₂O Emission: Pathways and Correlations

Sod-seeding patterns (Pr, Tr, TPr), as crucial agricultural management practices, have far-reaching impacts on soil nitrogen (N) cycling and nitrous oxide (N₂O) emissions, an important greenhouse gas. Here, the direct and indirect pathways through which sod-seeding patterns influence N₂O were analyzed, taking both abiotic (substrate availability) and biotic (microbial functional genes) factors into account. The sod-seeding pattern mainly decreased soil N₂O emission by reducing soil NH₄⁺-N content and nirS + nirK abundance. Soil NO₃⁻-N content was weakly affected by sod-seeding pattern but exerted strong influence on N₂O emission (Figure 6).

4. Discussion

4.1. Impacts on Soil Quality and Health

The integration of sod-seeding patterns (Pr, Tr, TPr) demonstrated profound effects on soil physicochemical properties, nutrient dynamics, and enzymatic activities, collectively enhancing soil quality and health in peach orchards.

The significant reduction in bulk density (most notable in TPr) and pH (approaching neutrality) across all treatments indicates improved soil structure and reduced compaction. Lower bulk density enhances root penetration and water infiltration, mitigating risks of waterlogging and root rot—a critical factor in peach cultivation where soil aeration is vital. The elevated electrical conductivity (EC) in all sod-seeded plots suggests enhanced solute mobility, potentially facilitating nutrient uptake by roots. Increased mean weight diameter (MWD) of soil aggregates (Figure 2d) reflects improved aggregate stability, which is essential for water retention and erosion resistance [29,30]. These changes align with practices such as deep tillage and organic mulching, which are known to optimize soil physical conditions.

Moreover, sod-seeding patterns significantly altered nutrient availability. The consistent increase in soil organic carbon (SOC) in Pr highlights its role in enhancing organic matter retention, a key driver of long-term soil fertility. The differential responses of nitrate (NO₃⁻-N) and ammonium (NH₄⁺-N) nitrogen further underscore treatment-specific nutrient dynamics: both TPr and Pr favored nitrate mobilization, while only Pr excelled in ammonium retention. Lolium perenn did not introduce new nitrogen into the soil like legumes do, but suppressed nitrification and slowed down the conversion of ammonium (NH₄⁺) to nitrate (NO₃⁻) [31], resulting in an accumulation of NH₄⁺ in the soil. This divergence may influence nitrogen use efficiency in peach trees, as nitrate is more mobile but prone to leaching [32,33], whereas ammonium is more stable but requires enzymatic conversion for plant uptake. The elevated soil organic nitrogen (SON) and total nitrogen (STN) in TPr and Tr suggest accelerated organic matter mineralization, potentially synchronizing nutrient release with peach growth phases. Notably, available phosphorus (SAP) peaked in Tr, emphasizing its efficacy in phosphorus bioavailability—a critical factor in calcareous soils where phosphorus fixation is common.

The surge in soil enzyme activities (SUC, URE, CAT) under sod-seeding treatments reflects enhanced microbial metabolic activity. Sucrase and urease, key enzymes in carbon and nitrogen cycling [34], were most elevated in TPr, indicating robust organic matter decomposition and nutrient mineralization [35]. Catalase activity across all treatments implies improved redox conditions, which could mitigate oxidative stress in soil microbes and promote a nitrification–denitrification balance. These enzymatic shifts correlate with microbial community restructuring, where sod-seeding likely enriched copiotrophic taxa capable of rapid organic matter turnover. The dominance of TPr in enzymatic kinetics suggests its superiority in fostering a dynamic microbial ecosystem, which is critical for sustaining nutrient fluxes under intensive orchard management.

In summary, sod-seeding patterns comprehensively improved the soil quality through optimizing physicochemical properties, nutrient supply, and enzyme activity. The soil quality improvement underscored the dominance of TPr in enzyme kinetics and nutrient mobilization and the ascendancy of Pr in organic matter retention and pH moderation.

4.2. Pathways from Sod-Seeding Pattern to Nitrogen Substrates and Functional Genes

The sod-seeding pattern showed a significant negative correlation with soil ammonium (NH₄⁺-N) concentration [36], suggesting that different seeding strategies, such as single-cropping (Pr, Tr) or intercropping (TPr), can influence NH₄⁺ dynamics through processes such as enhanced microbial immobilization, increased plant N uptake, or reduced ammonification [37]. For instance, intercropping (TPr) might make use of complementary resources among different species, speeding up the consumption of NH₄⁺ compared to its production, thereby reducing the soil NH₄⁺ pools.

The sod-seeding pattern had a significant negative impact on the combined abundance of nirK and nirS genes, which encode nitrite reductases crucial for denitrification by converting nitrite (NO₂⁻) to N₂O or dinitrogen (N₂) [38,39]. A decrease in these genes under specific seeding patterns indicates changes in the denitrifier community structure, resulting in reduced potential for N₂O production during denitrification [40].

4.3. Direct Effects of Nitrogen Substrates and Functional Genes on N₂O

Soil NH₄⁺-N had a strong positive correlation with N₂O emissions. Higher NH₄⁺ availability directly promotes N₂O formation by increasing the substrate supply for N-transforming microbes [41]. Primarily, NH₄⁺-N is the substrate for autotrophic nitrification, the two-step oxidation of ammonia to nitrate, which is a direct source of N₂O [42]. Elevated concentrations of NH₄⁺-N can stimulate the growth and activity of ammonia-oxidizing microorganisms, including both ammonia-oxidizing bacteria (AOB) and archaea (AOA) [43]), thereby increasing the overall rate of nitrification and associated N₂O leakage.

Furthermore, high NH₄⁺-N availability can promote N₂O emissions through indirect pathways. Under conditions of high substrate concentration, rapid nitrification can lead to the accumulation of nitrite (NO₂⁻), a highly reactive intermediate. This accumulation can fuel N₂O production via nitrifier denitrification, a process where ammonia oxidizers reduce NO₂⁻ to N₂O, particularly under low oxygen conditions [13]. Additionally, the nitrate (NO₃⁻) produced from nitrification serves as the terminal electron acceptor for canonical denitrification, another major source of N₂O [44]. Thus, a high supply of NH₄⁺-N effectively fuels the entire coupled nitrification–denitrification sequence, leading to amplified N₂O emissions. The strong correlation observed in our study likely reflects the integrated contribution of these multiple, interconnected pathways, all of which are fundamentally driven by the availability of NH₄⁺-N.

The combined abundance of nirK + nirS genes also had a significantly positive relationship with N₂O emission. The nirK and nirS genes encode enzymes that convert nitrite (NO₂⁻) to nitric oxide (NO), a crucial step in denitrification. The total gene count in a soil sample reflects the community’s ability to perform this conversion. More nirK + nirS genes lead to higher N₂O emissions, especially in low-oxygen conditions (a characteristic of denitrification hotspots) [45,46]. This direct link emphasizes how the metabolic potential of microbes determines N₂O emissions [47].

Numerous studies in various ecosystems and management practices support this finding. Agricultural activities that boost N₂O emissions, like nitrogen fertilization, also increase nirK and nirS gene numbers [15,48]. Conversely, practices reducing N₂O, such as certain biochar applications, can alter the denitrifying community by reducing nirK + nirS gene levels or changing their ratio to other denitrification genes [49,50]. This consistent pattern underscores the key role of nitrite-reducing microbes in regulating N₂O production potential.

For nitrate (NO₃⁻-N), sod-seeding pattern showed a non-significant positive pathway (coefficient = 0.39, p > 0.05), indicating a weak direct regulation of the sod-seeding pattern on soil nitrate pools. However, NO₃⁻-N had a strong negative correlation with N₂O emissions [51]. This negative association is in line with the “denitrification completion” hypothesis: sufficient NO₃⁻-N promotes complete denitrification (converting NO₃⁻ to N₂), reducing the accumulation of the intermediate N₂O [52]. Alternatively, high NO₃⁻ may cause changes in the microbial community that favor complete denitrification over incomplete pathways, limiting the release of N₂O [53].

4.4. Integrated Mechanisms and Implications

Sod-seeding pattern affects N₂O through multiple, interconnected pathways:

(1)

Substrate-driven pathway: The pattern regulates NH₄⁺-N, which directly fuels N₂O production by providing substrates for nitrification and denitrification [51].

(2)

Microbial-functional pathway: The pattern controls the abundances of nirK + nirS, changing the denitrification potential and N₂O formation [54].

(3)

Competitive-denitrification pathway: The pattern has a weak direct effect on NO₃⁻-N but strongly influences N₂O through NO₃⁻-mediated completion of denitrification, reducing the intermediate N₂O [55].

These results showed that the sod-seeding pattern shaped N₂O emissions through both abiotic (substrate availability) and biotic (microbial gene regulation) mechanisms. For N₂O mitigation in sod-seeded agroecosystems, strategies should aim to reduce NH₄⁺ accumulation, for example, through enhanced plant uptake or microbial immobilization and modulate denitrifier functional genes, such as through species intercropping that favors complete denitrification.

4.5. Implications, Limitations, and Future Directions

The integration of sod-seeding patterns (Pr, Tr, TPr) demonstrates significant potential to reduce N₂O emissions while enhancing soil health in peach orchards. By mimicking natural ecosystems through grass–legume integration (e.g., Tr and TPr), these systems reduce reliance on synthetic fertilizers, aligning with conservation agriculture principles [56]. For instance, Pr’s ability to retain soil organic carbon (SOC) and moderate pH mitigates soil degradation and carbon sequestration losses, which is critical for long-term fertility [57]. TPr’s dominance in enzymatic activity (e.g., SUC, URE) enhances organic matter decomposition, promoting nutrient synchronization with peach growth phases and buffering against seasonal fluctuations. This enzymatic resilience, coupled with reduced bulk density and improved aggregate stability (MWD), mitigates erosion risks and optimizes water retention—key adaptations to climate variability. Moreover, TPr’s capacity to lower NH₄⁺-N pools and modulate denitrifier functional genes (nirK/nirS) suppresses N₂O production pathways, while elevated NO₃⁻-N under Tr facilitates complete denitrification, further reducing intermediate N₂O accumulation. These findings position sod-seeding as a dual-functional strategy for productivity and ecological resilience.

Despite promising outcomes, current studies face critical limitations. First, short-term trials (e.g., 3–5 years) may overlook the long-term impacts on microbial community dynamics and SOC sequestration. For example, Pr’s initial SOC gains might plateau over time due to decomposition dynamics, necessitating extended monitoring [58,59]. Second, variability in treatment responses (e.g., SON decline in Pr) highlights unresolved mechanistic interactions between plant–microbe–soil systems. The reduced SON in Pr suggests potential trade-offs between organic matter retention and nutrient cycling efficiency, requiring deeper metagenomic profiling to elucidate the microbial taxa responsible. Third, regional adaptability remains unvalidated. While TPr excels in nutrient mobilization in calcareous soils, its efficacy in waterlogged or saline-sodic soils—common in peach-growing regions—requires validation through multi-location trials. Lastly, economic feasibility studies are lacking. Although TPr reduces N-fertilizer inputs, the cost–benefit ratio of sod-seeding implementation (e.g., seedling establishment and maintenance are more labor-intensive than clean tillage for the first year) remains unquantified, limiting adoption by resource-limited farmers.

To advance sod-seeding applications in orchards, future research should prioritize three axes. First, longitudinal studies spanning 10 years or more are needed to assess SOC dynamics, microbial succession, and N₂O mitigation efficacy under climate extremes (e.g., droughts, heavy rainfall). Isotopic tracing (e.g., ¹⁵N) could quantify N transformation pathways [60] and identify keystone microbial guilds driving denitrification. Second, integrating sod-seeding with precision agriculture tools (e.g., sensor-guided fertilization, drones for soil mapping) could optimize resource allocation. For instance, coupling TPr with nitrate sensors might balance N supply–demand while minimizing leaching. Third, expanding functional genomics to profile nirK/nirS-harboring taxa will clarify how sowing patterns reshape denitrifier communities. Metatranscriptomic analyses could reveal metabolic shifts under varying moisture and O₂ regimes [61], guiding gene-editing strategies to enhance complete denitrification. By bridging ecological principles with technological innovation, sod-seeding patterns can evolve into a scalable paradigm for climate-smart agriculture.

5. Conclusions

The sod-seeding patterns (Pr, Tr, TPr) delivered dual benefits for peach orchard sustainability: mitigating nitrous oxide (N₂O) emission and enhancing soil health. TPr (Trifolium repens–Lolium perenne mixed sowing) excelled, reducing cumulative N₂O emission by 34.0 to 41.0% via suppressing NH₄⁺-N pools and nirK/nirS genes—key drivers of denitrification. The recommended seed amounts for Trifolium repens and Lolium perenne in TPr pattern are 0.5 and 1 kg·667 m⁻², respectively. All sod-seeding treatments improve soil physicochemical properties, optimize nutrient dynamics, and stimulate enzymatic activities. These changes advance soil resilience, nutrient cycling, and N₂O mitigation, positioning sod-seeding as a climate-smart strategy for long-term orchard productivity and ecological balance.