Glucose Elevates N₂O Emissions by Promoting Fungal and Incomplete Denitrification in North China Vegetable Soils

Abstract

Agricultural soils are hotspots of nitrous oxide (N₂O) emissions, where carbon substrates act as a critical factor influencing microbial community composition. However, how carbon availability modulates microbial denitrifying pathways and further influences N₂O emissions remains poorly understood. Here, we conducted anaerobic incubations to investigate North China vegetable soil N₂O production and consumption in response to varied glucose concentrations (0, 0.5 (Glu_0.5), 1.0 (Glu_1.0), and 2.0 (Glu_2.0) g C kg⁻¹ d.w. of soil). In this study, the δ¹⁵N^SP/δ¹⁸O mapping approach (δ¹⁵N^SP/δ¹⁸O MAP) and acetylene inhibition technique (AIT) were used to quantify the residual N₂O ratio (r_N2O) and the relative contributions of bacterial (f_BD) and fungal (f_FD) denitrification to N₂O production. The results showed that increasing glucose concentrations significantly increased CO₂ and N₂O emissions, with peak fluxes observed at Glu_2.0 on day 1 (116.22 ± 2.80 mg CO₂-C kg⁻¹ and 1.08 ± 0.02 mg N₂O-N kg⁻¹). Concurrently, δ¹⁵N^SP was also significantly elevated (p < 0.001), indicating enhanced f_FD, which was further corroborated by positive correlations between f_FD and glucose concentration (r = 0.48–0.56, p < 0.001). Nevertheless, bacterial denitrification (BD) still dominated N₂O production throughout the incubation period, except on day 1 in Glu_1.0 and Glu_2.0 of Case 2. Bland–Altman analysis with 95% limits of agreement (LoA) demonstrated strong agreement between the MAP and AIT in r_N2O estimation, particularly under Glu_2.0. All the above revealed glucose-induced denitrifying microbial shifts from BD to fungal denitrification (FD), which consequently modulated N₂O emissions and promoted incomplete denitrification. These findings collectively demonstrate that in vegetable cropping systems, rational carbon management strategies can promote N₂O reduction to N₂, thereby achieving effective N₂O mitigation.

1. Introduction

Nitrous oxide is a potent greenhouse gas, contributing to global warming and stratospheric ozone depletion [1,2]. Arable soils (3.9 Tg N yr⁻¹ in 2020) are the largest source of anthropogenic N₂O emissions (6.7 Tg N yr⁻¹ in 2020), where the excessive use of synthetic nitrogen fertilizers has substantially exacerbated N₂O emissions [2,3,4]. For fertilizer nitrogen-induced N₂O emissions in arable soils, carbon source is a key factor that affect soil conditions and further influences N₂O emissions by modulating its microbial pathways (e.g., nitrification, bacterial and fungal denitrification) [5,6,7,8].

Denitrification is a facultative anaerobic process mediated by heterotrophic denitrifying microorganisms (bacteria and fungi), involving the stepwise reduction of ${\mathrm{N}\mathrm{O}}_3^{-}$ to ${\mathrm{N}\mathrm{O}}_2^{-}$, NO, N₂O, and ultimately N₂. During this process, the terminal reductive conversion of N₂O to N₂ may be incomplete, leading to N₂O to be released into the atmosphere. This conversion depends on microbial community composition and environmental conditions [9]. Notably, fungal and bacterial denitrification exhibit distinct enzymatic mechanisms. This results in bacterial denitrification (BD) being able to undergo a complete denitrification process and emit N₂O as intermediate products [5], whereas fungal denitrification (FD) emits N₂O as an obligatory end product due to the absence of N₂O reductase [10,11]. The quality of carbon substrates, an important abiotic factor regulating this heterotrophic microbial nitrogen (N) process, is known to shape fungal and bacterial growth and activity differently [12,13,14]. Although numerous studies have elucidated N₂O production mechanisms under varying soil environmental conditions [7], or at microbial scales [13], research on how carbon availability regulates microbial pathways of N₂O production and its impact on emission magnitude remains scarce.

Soil carbon (C) availability is a key factor of regulating denitrification rates, given that labile C serves as the primary electron donor for each reductive step in the conversion of ${\mathrm{N}\mathrm{O}}_3^{-}$ to N₂ [15]. Incorporating organic residues into agricultural soils is commonly recommended to improve soil fertility and to increase soil carbon content for mitigating global warming [16,17]. Animal manures and crop residuals are the main sources of organic carbon compounds in these soils. Numerous studies have tested the effect of crop residuals on denitrifying N₂O production and residual N₂O rates (r_N2O) [18,19,20]. Wu et al. [19] conducted an incubation experiment to investigate the combined effect of rice straw and KNO₃ application on soil N₂O and N₂ fluxes under variable O₂ partial pressure. They found that the combined rice straw and KNO₃ can increase soil N₂O emissions under conditions favoring denitrification, but it decreases the overall r_N2O. Moreover, Senbayram et al. [18] suggested that in a sandy soil amended with maize straw, KNO₃ is the primary factor that regulating r_N2O of denitrification. Heterotrophic bacteria and fungi are primary decomposers of organic matter in agroecosystems; however, they exhibit differences in competitive ability for resource exploitation. It is generally believed that fungi could outgrow bacteria in utilizing complex carbon compounds (e.g., cellulose and lignin), whereas bacteria prefer simple compounds (e.g., glucose). Soil environmental conditions (e.g., soil pH, concentration of available carbon) can also affect fungal and bacterial ability in utilizing carbon sources [21,22]. Fungi are more competitive than bacteria in utilizing carbon compounds under acidified soil conditions and tend to produce more N₂O [21]. Furthermore, they exhibit even stronger competitiveness when the soil is amended with a high concentration of glucose (e.g., >10 mg C g⁻¹ d.w. of soil) for a short period [22]. All the above may be attributed to that fungi appear to be more adaptable to adverse environmental conditions than bacteria. Taken together, the relationships among carbon availability levels, the fractions of denitrifying bacterial (f_BD) and fungal (f_FD) producing N₂O, and r_N2O are complex. Consequently, for vegetable soils with high fertilizer demands, elucidating these interrelationships is critical for developing targeted N₂O mitigation strategies. However, how the fungal and bacterial competitive ability for carbon source acquisition varies with the level of carbon available remains unclear.

In this study, we hypothesized that glucose availability would differentially shape the contributions of fungi and bacteria to N₂O production from denitrification, consequently altering the N₂O production–reduction balance in a carbon-dependent manner. To test this hypothesis, we conducted anaerobic incubation experiments with North China arable soil. Our objective was to assess the effect of glucose concentration on denitrifying N₂O production and consumption. Furthermore, to elucidate the underlying microbial mechanisms involved, the relative bacterial and fungal contributions to N₂O production from denitrification as well as the residual N₂O ratios were investigated by determining the isotope signatures of the N₂O produced.

2. Materials and Methods

2.1. Site Description and Soil Properties

Soil samples were collected from a vegetable field site where cabbages have been planted successively for seven years at the Environmental Research Station (40°15′ N, 116°55′ E) of the Chinese Academy of Agricultural Sciences located in Beijing, China. The soil type is classified as calcareous fluvo-aquic (FAO Classification) [23,24]. The experimental soil samples were collected from the upper 20 cm after removing the top 2 cm layer of plant residuals and soil (See Supplementary Text S1 for details). After that, soil samples were sieved through 4 mm mesh and air dried. The soil had a pH value of 7.8 and a bulk density of 1.3 g cm⁻³ and contained 11.4 g kg⁻¹ of organic C and 1.1 g kg⁻¹ of total N. Soil initial ${\mathrm{N}\mathrm{H}}_4^{+}$-N and ${\mathrm{N}\mathrm{O}}_3^{-}$-N contents were 4.0 and 242.6 mg N kg⁻¹, respectively.

2.2. Experimental Design

To initiate and stabilize the microbial activity prior to applying fertilizers, the soil was pre-incubated in a constant temperature incubator at 25 °C and maintained at 37% water-filled pore space (WFPS) for two weeks [23,25]. The experiment comprised four glucose concentration levels (

Table 1. Glucose concentrations and acetylene variants.

Factor	Variants
Glucose gradients	0	0.5	1.0	2.0
(g C kg−1 d.w. of soil)	(Control)	(Glu_0.5)	(Glu_1.0)	(Glu_2.0)
Acetylene variants	0		10
(kPa)	Non-acetylene-treated set		Acetylene-treated set

): (1) non-amended control (Control); (2) soil with the addition of glucose at 0.5 g glucose-C kg⁻¹ dry weight of soil (d.w. of soil; Glu_0.5); (3) soil with the addition of glucose at 1.0 g glucose-C kg⁻¹ d.w. of soil (Glu_1.0); (4) soil with the addition of glucose at 2.0 g glucose-C kg⁻¹ d.w. of soil (Glu_2.0). Each level included both an acetylene-treated set (10 kPa) and a non-acetylene-treated set (0 kPa, without acetylene), resulting in a total of eight distinct treatments (n = 4). An equivalent of 60 g d.w. of soil was packed into each flask (250 cm³) at a bulk density of 1.2 g cm⁻³. KNO₃ (100 mg N kg⁻¹ d.w. of soil) and glucose were applied in the form of solutions, with deionized water added to achieve a final moisture content of 80% WFPS across all treatments. The unamended control received only KNO₃ solution (without glucose) and deionized water.

On sampling days (1, 2, 3, 4, 6, and 8), all flasks were sealed with stoppers. They were then evacuated and refilled with pure helium (99.995% purity, He) for three cycles to remove atmospheric air from soil pores and headspace. Following this, flasks underwent a final evacuation/He refill cycle to establish anaerobic conditions favorable for denitrification. For acetylene-treated flasks, 25 cm³ of headspace He was replaced with an equal volume of acetylene (99.6% purity) using gastight syringes. All samples were incubated at 25 °C prior to gas sampling. After 2 h incubation, two 30 cm³ of headspace gas were collected into pre-evacuated vials for N₂O emission rate and isotopic composition analysis, respectively. Soil moisture was maintained at constant level by weighing flasks every other day (day 2, 4, 6, and 8) and replenishing distilled water as needed to compensate for evaporation.

2.3. Gas Analysis

The N₂O concentration was determined using a gas chromatograph (GC; Agilent 7890A, Santa Clara, CA, USA) equipped with an electron capture detector (ECD), and the emission rate of N₂O was then calculated based on the equation described by Zheng et al. [5].

The isotope signatures (δ¹⁵N^bulk, δ¹⁵N^α, and δ¹⁸O) of N₂O were determined using a Delta V Plus isotope ratio mass spectrometer (Thermofisher, Bremen, Germany) by the detection of mass-to-charge ratio (m/z) 44, 45, and 46 of N₂O+ molecular ions and m/z 30 and 31 of NO+ fragment ions [26,27].

2.4. Soil Analysis

Surface soil (0–20 cm) was collected with gas samples and subsequently oven-dried at 105 °C for 24 h to measure water-filled pore space (WFPS). Soil pH was measured in soil slurry (1:2.5, d.w. of soil:distilled water) using a Thermo Orion pH meter (Mettler Toledo, Shanghai, China). The soil mineral N was extracted with 2 M KCl (1:5, d.w. of soil:KCl solution), and their contents were determined by a continuous flow analyzer (QuikChem8000, LACHAT, Loveland, CO, USA). Soil organic carbon was determined using an Elemental Analyzer (Flash 2000; Thermo Fisher Scientific, Waltham, MA, USA).

2.5. N₂O Source Partitioning

Two-isotope plots of ¹⁵N site preference (δ¹⁵N^SP = 2 × δ¹⁵N^α − 2 × δ¹⁵N^bulk) of the produced N₂O vs. δ¹⁸O (δ¹⁵N^SP/δ¹⁸O mapping approach, δ¹⁵N^SP/δ¹⁸O MAP), introduced by Lewicka-Szczebak et al. [28], have recently been used to simultaneously calculate the N₂O reduction faction and the N₂O fraction from bacterial denitrification in both field and laboratory studies [6,8,28,29,30,31,32]. The δ¹⁸O values of N₂O are largely/substantially influenced by the oxygen exchange between substrate nitrate and water [29,33]. Thus, endmember δ¹⁸O in the MAP needs to be corrected based on the δ¹⁸O values of substrates ${\mathrm{N}\mathrm{O}}_3^{-}$ and H₂O, along with their oxygen exchange ratio observed in this study (see Table S1 in the Supplement). The MAP involves an endmember mixing model for bacterial fraction and the Rayleigh model for N₂O reduction fraction [29]. Notably, N₂O production processes are assumed to occur under the following two scenarios: in scenario i, N₂O produced by BD is first partially reduced to N₂, and the residual N₂O is then mixed with that produced by FD (Case 1: R-M, Equation (1)); in scenario ii, N₂O produced by BD and FD is first mixed, and the mixed N₂O is then reduced to N₂ (Case 2: M-R, Equation (2)) [29].

$${\mathsf{\delta }}_{\mathrm{sample}}={(\mathsf{\delta }}_{\mathrm{BD}}{+{\mathsf{\epsilon }}_{\mathrm{r}}\ln r_{\mathrm{N}2\mathrm{O}})f}_{\mathrm{BD}}+{\mathsf{\delta }}_{\mathrm{FD}}(1-f_{\mathrm{BD}})$$

(1)

$${\mathsf{\delta }}_{\mathrm{sample}}=\left[{\mathsf{\delta }}_{\mathrm{BD}}{f}_{\mathrm{BD}}+{\mathsf{\delta }}_{\mathrm{FD}}(1-f_{\mathrm{BD}})\right]+{\mathsf{\epsilon }}_{\mathrm{r}}\ln r_{\mathrm{N}2\mathrm{O}}$$

(2)

where δ_sample refers to δ¹⁸O or δ¹⁵N^SP of N₂O. f_BD is the N₂O fraction from BD. δ_BD and δ_FD represent the endmember isotopic values of BD and FD, respectively, in the dual-isotope plot, adopted from the literature. ε_r denotes the net isotope effect of ¹⁸O or ¹⁵N^SP during the reduction of N₂O to N₂. r_N2O indicates the N₂O residual ratio.

Moreover, r_N2O can also be determined using the acetylene inhibition techniques (AIT). We use the measured [N₂O] in the non-acetylene-treated group, [N₂O]_residual, and that in the acetylene-treated group, [N₂O + N₂], to calculate the residual fraction of N₂O (r_N2O) following Equation (3) [5].

$$r_{{\mathrm{N}}_2\mathrm{O}}={{[\mathrm{N}}_2\mathrm{O}]}_{\mathrm{residual}}/{[\mathrm{N}}_2\mathrm{O}+{\mathrm{N}}_2]$$

(3)

2.6. Statistical Analysis

The impact of glucose concentration on CO₂ and N₂O emissions, N₂O isotope signatures, residual N₂O ratio (r_N2O), and the bacterial (f_BD) and fungal (f_BD) contributions to denitrifying N₂O production was assessed using the non-parametric Kruskal–Wallis test. Post-hoc pairwise comparisons were conducted using Wilcoxon rank-sum tests, with p-values adjusted via the Benjamini–Hochberg (BH) method to control for multiple testing errors. Pearson’s correlation coefficient (r) was used to assess the correlations among the incubation time, glucose concentration, CO₂, N₂O, N₂O isotope signatures, r_N2O, f_BD, and f_FD. Statistical significance was set at p < 0.05. The agreement between the δ¹⁵N^SP/δ¹⁸O MAP and the acetylene inhibition technique for estimating r_N2O was evaluated using Bland–Altman analysis with 95% limits of agreement (LoA). Statistical differences in r_N2O between the two methods were assessed using paired t-tests.

3. Results

3.1. CO₂ and N₂O Emissions in Response to Varied Glucose Concentrations

For all tested glucose concentrations, daily CO₂ emission patterns exhibited consistent temporal trends, characterized by a general decline. The concentration gradients of glucose significantly influenced CO₂ emission rates (p < 0.001; Figure 1c,f), with the 2.0 g glucose-C kg⁻¹ dry weight of soil (d.w. of soil; Glu_2.0) treatment yielding the highest average daily flux of 118.06 ± 7.29 mg CO₂-C kg⁻¹ d.w. of soil in the acetylene-treated group (Figure 1b) and 116.22 ± 2.80 mg CO₂-C kg⁻¹ d.w. of soil in the non-acetylene-treated group (Figure 1a) on day 1. This value was 36.07- and 35.08-fold higher than that in the Control (without glucose), respectively. This was further supported by Pearson coefficient correlation analysis, which showed a significantly positive correlation between glucose concentration and daily CO₂ emissions (both the acetylene-treated and non-acetylene-treated groups showed r = 0.43, p < 0.001; Figure 2). Whereas no obvious changes in daily CO₂ emissions were observed with the addition of acetylene (Figure 1c).

Similar to CO₂ emission patterns, increasing glucose concentrations significantly enhanced daily N₂O emission rates (p < 0.001; Figure 1d–f). In the non-acetylene-treated group, peak N₂O emissions increased from (2.81 ± 0.29) × 10⁻³ mg N₂O-N kg⁻¹ d.w. of soil on day 3 in the Control to 1.08 ± 0.02 mg N₂O-N kg⁻¹ d.w. of soil on day 1 in Glu_2.0 treatment (Figure 1d). Whereas, in the acetylene-treated group, peak emissions rose from (2.68 ± 1.84) × 10⁻² mg N₂O-N kg⁻¹ d.w. of soil on day 1 in the Control to 3.75 ± 1.19 mg N₂O-N kg⁻¹ d.w. of soil on day 3 in Glu_2.0 treatment (Figure 1e). Positive correlations between glucose concentration and daily N₂O emissions further supported this pattern (r = 0.63, p < 0.001 in the non-acetylene-treated group and r = 0.58, p < 0.001 in the acetylene-treated group; Figure 2). Notably, acetylene addition significantly increased daily N₂O emission rates compared to the non-acetylene-treated group, particularly in the Control (p < 0.0001) and Glu_2.0 (p < 0.001) treatments (Figure 1f).

3.2. N₂O Isotope Signatures in Response to Varied Glucose Concentrations

The isotope signatures (δ¹⁸O and δ¹⁵N^SP) of N₂O produced by soils amended with 0.5 (Glu_0.5), 1.0 (Glu_1.0), and 2.0 g glucose-C kg⁻¹ d.w. of soil were analyzed in conjunction with flux measurements to identify N₂O production pathways (Figure 3a–f). Isotopic signatures of N₂O in the unamended control group were undetectable, as emission fluxes fell below the analytical detection limit.

3.2.1. δ¹⁵N^SP of N₂O

For all glucose gradients, δ¹⁵N^SP values of N₂O exhibited significant temporal variation throughout the incubation period in both the non-acetylene-treated and acetylene-treated groups (p < 0.01; Figure 3d–f). This temporal pattern was supported by significant negative correlations between incubation duration and δ¹⁵N^SP values (r = −0.63 to −0.62, p < 0.001; Figure 3a,b). Notably, δ¹⁵N^SP values were highest on day 1, ranging from 16.12 ± 0.14‰ for Glu_0.5 treatment to 25.11 ± 0.04‰ for Glu_1.0 treatment to 33.02 ± 3.44‰ for Glu_2.0 treatment, and showed temporal patterns consistent with N₂O emission dynamics, exhibiting consecutive depletion during the initial three days for all treatments except the Glu_0.5 group. However, these values changed minimally and showed no clear pattern over the subsequent incubation period. Increasing glucose concentrations significantly elevated δ¹⁵N^SP values of N₂O in both the non-acetylene-treated and acetylene-treated groups (p < 0.001; Figure 3f), which was further supported by significant positive correlations between glucose concentration and δ¹⁵N^SP value (r = 0.47 to 0.76, p < 0.001; Figure 3a,b). Furthermore, acetylene addition significantly depleted δ¹⁵N^SP values of both Glu_1.0 and Glu_2.0 (p < 0.001), resulting in a decrease of 2.48‰ to 12.79‰ compared to the non-acetylene-treated group.

3.2.2. δ¹⁸O of N₂O

Consistent with δ¹⁵N^SP trends, δ¹⁸O values of N₂O peaked on day 1 in both the non-acetylene-treated (44.62 ± 0.33‰ for Glu_1.0 group and 49.27 ± 0.44‰ for Glu_2.0 group) and acetylene-treated (28.91 ± 0.91‰ for Glu_1.0 group and 34.44 ± 0.23‰ for Glu_2.0 group) groups (Figure 3a,b). These δ¹⁸O values then progressively depleted during the initial three days (p < 0.05) for all treatments except the Glu_0.5 group. Different from δ¹⁵N^SP, δ¹⁸O values of N₂O showed no significant variation across glucose concentrations in either non-acetylene-treated (p = 0.316) or acetylene-treated (p = 0.256) groups (Figure 3c). Moreover, acetylene addition significantly depleted δ¹⁸O values at all glucose concentrations (p < 0.001), with reductions ranging from 5.44‰ to 19.59‰ compared to the corresponding non-acetylene-treated groups.

3.3. Denitrifying Bacterial and Fungal Contributions to N₂O Production in Response to Varied Glucose Concentrations

The measured δ¹⁵N^SP and δ¹⁸O values in the non-acetylene-treated group were used as input for the MAP model. The majority of the data points were distributed within the assumed boundaries defined by mixing and reduction lines. Only two were located outside the boundaries. For these samples, the calculation results provide f_BD and r_N2O were slightly above 1, which were set to 1 for the further analyses. Figure 4 showed that samples in the non-acetylene-treated group were closer to the reduction line compared to those in the acetylene-treated group. However, in both groups, samples treated with higher glucose concentrations were further away from the bacterial endmember. This indicated that the bacterial contribution to denitrifying N₂O production (f_BD) decreased with increasing glucose concentrations, a result that has also been quantitatively confirmed through the mapping model.

The calculation has been performed with two cases: Case1: R-M and Case 2: M-R (see Section 2.4). Under both scenarios in this study, the increase in glucose concentrations significantly promoted f_FD (p < 0.001; Figure 5d1,d2). Compared to the R-M scenario, the M-R scenario yielded higher f_FD under all glucose-amended treatments (Figure 5a1–c1,a2–c2). Nevertheless, BD still dominated N₂O emissions from the tested soil (f_BD range: 0.52 ± 0.03 to 1 ± 0), except on day 1 under Glu_1.0 (0.38 ± 0) and Glu_2.0 (0.17 ± 0.11) in Case 2.

3.4. N₂O Residual Rates in Response to Varied Glucose Concentrations

Both the δ¹⁵N^SP/δ¹⁸O MAP and the acetylene inhibition technique (AIT) were used to determine the N₂O residual ratio (r_N2O). The results indicated that increasing glucose concentrations significantly increased r_N2O for both cases (p < 0.05 for Case1 (r_N2O_case1) and p < 0.001 for Case2 (r_N2O_case2); Figure 5d1,d2), which was corroborated with Spearman correlations between r_N2O_case1 and glucose concentration (r = 0.33, p < 0.01) and between r_N2O_case2 and glucose concentration (r = 0.56, p < 0.001; Figure 6). As shown in Figure 5a1,a2, for the Glu_0.5 treatment, r_N2O peaked on day 1 in both cases (0.37 ± 0.01 for r_N2O_case1 and 0.44 ± 0.01 for r_N2O_case2). This was followed by a temporary decline; thereafter, r_N2O exhibited an overall decreasing trend with fluctuations. When the glucose concentration increased to Glu_1.0, the r_N2O peak was delayed to day 3 (0.40 ± 0.02 for r_N2O_case1 and 0.47 ± 0.02 for r_N2O_case2; Figure 5b1,b2), with subsequent temporal trends similar to those observed in the Glu_0.5 treatment. With a further increase in glucose concentration to Glu_2.0, r_N2O progressively increased over the initial three days (Figure 5c1,c2), reaching its first peak on day 3 (0.46 ± 0.05 for r_N2O_case1 and 0.52 ± 0.04 for r_N2O_case2). This was followed by a sudden decrease (p < 0.05), after which r_N2O progressively increased again over the final four days, reaching its highest value on day 8 in both cases (0.95 ± 0.09 for r_N2O_case1 and 0.95 ± 0.08 for r_N2O_case2).

Comparing the r_N2O determined by the AIT (r_N2O_AIT) with that determined by the δ¹⁵N^SP/δ¹⁸O MAP (r_N2O_case1 and r_N2O_case2) showed that the temporal trends of r_N2O were similar between the two methods (Figure 7a). Although no significant difference was observed for r_N2O_AIT between Glu_1.0 and Glu_2.0 (p = 0.346), the average r_N2O_AIT was slightly higher in the Glu_1.0 treatment than in the Glu_2.0 treatment, which differed from the results of the MAP (Figure 7b). Bland–Altman analysis was performed to evaluate the concordance between MAP and AIT in r_N2O estimation. For the Glu_2.0 treatment, results revealed nonsignificant mean biases of −0.048% (95% LoA: −0.496% to 0.400%) for r_N2O_case1 vs. r_N2O_AIT (Figure 8c1) and 0.065% (95% LoA: −0.258% to 0.388%) for r_N2O_case2 vs. r_N2O_AIT (Figure 8c1). For the two other treatments, significant mean biases were observed for both comparisons. Nevertheless, nearly all data points across all treatments fell within the 95% limits of agreement (Figure 8a1,a2,b1,b2). The Bland–Altman analysis revealed good consistency between the r_N2O values assessed by the two methods.

4. Discussion

4.1. Denitrifying Mechanisms Driven by Varied Glucose Concentrations

Based on the R-M scenario of the δ¹⁵N^SP/δ¹⁸O MAP, increased glucose concentration promoted complete denitrification by enhancing both the reductive conversion of N₂O to N₂ and f_BD during the initial incubation period (on day 1). Throughout the incubation period, however, increased glucose concentrations inhibited complete denitrification, leading to more residual N₂O emitted into the atmosphere. This contrasts with previously reported findings that organic carbon amendment stimulates complete denitrification and mitigates N₂O emissions [16,20]. In this study, the promoting effect of glucose on N₂O emissions was primarily attributed to enhanced fungal denitrification. The microbial community mediating this process lacks functional enzymes catalyzing the reduction of N₂O to N₂ [11], resulting in N₂O accumulation. Moreover, glucose failed to induce nosZ expression (which governs N₂O reduction to N₂) in the denitrifiers [34], favoring N₂O accumulation over reduction. Based on AIT, all glucose-amended treatments exhibited inhibited complete denitrification compared to the unamended control. This contrasts with our previous study [5], where the 300 mg glucose-C kg⁻¹ d.w. of soil amendment enhanced complete denitrification compared to the unamended control. This discrepancy may be attributed to the fact that the relatively lower glucose concentration in our prior study, which failed to induce a significant enhancement of fungal contribution to N₂O emission from denitrification.

For soils amended with Glu_0.5, f_BD increased progressively as incubation time progressed (p < 0.001). This trend was likely attributable to the gradual depletion of carbon resources over increasing incubation time. Under such conditions, limited carbon resources were preferentially utilized by denitrifying bacteria over fungi, thereby promoting bacterial denitrification. This finding demonstrates that bacteria are more competitive than fungi in utilizing simple carbon sources, which aligns with previous studies regarding the effects of different carbon sources (glucose, cellulose, winter pea, and switchgrass) on N₂O emissions [13]. When glucose concentration increased to Glu_1.0, compared to the Glu_0.5 treatment, as incubation time progressed, elevated carbon availability intensified competition between FD and BD over the course of incubation. By the end of the incubation period, this competition ultimately resulted in enhanced bacterial denitrification, as denitrifying bacteria outcompeted fungi for simple carbon resources [13]. With glucose concentration further increased to Glu_2.0, f_FD increased, whereas f_BD decreased with prolonged incubation except for an extremely high f_BD observed on day 1 in Case 2. Compared to the two other glucose-amended treatments, this opposite trend may be attributed to glucose-induced osmotic pressure inhibiting bacterial growth in the tested soil [22]. This f_FD enhancement induced by increasing glucose concentrations was corroborated by both a significant increase in δ¹⁵N^SP with rising glucose concentrations (p < 0.001) and a significant positive correlation between f_FD and glucose concentration (Case 1: r = 0.48, p < 0.001; Case 2: r = 0.56, p < 0.001; Figure 6).

4.2. Agreement Assessment of r_N2O Estimated by the δ¹⁵N^SP/δ¹⁸O MAP and AIT

An effective inhibition in the N₂O reduction to N₂ was confirmed by the significant depleted δ¹⁵N^SP and δ¹⁸O values of N₂O with the addition of acetylene [35,36,37]. Therefore, an agreement assessment of r_N2O estimated by the δ¹⁵N^SP/δ¹⁸O MAP and AIT can be further conducted. In treatments amended with lower glucose concentrations (Glu_0.5 and Glu_1.0), significant mean biases were consistently observed when comparing r_N2O derived from the δ¹⁵N^SP/δ¹⁸O MAP under both scenarios to those obtained by the AIT. Conversely, in the treatment amended with Glu_2.0, the same comparisons showed non-significant mean biases. This discrepancy was likely attributable to the increased glucose concentrations promoting f_FD, resulting in a greater proportion of N₂O escaping reduction to N₂. This mechanism consequently improved the agreement of r_N2O comparisons between the two methods. Furthermore, this interpretation is corroborated by the significant positive correlations observed between r_N2O values derived from both MAP scenarios and those obtained by the AIT (r_N2O_case1 vs. r_N2O_AIT: r = 0.44, p < 0.001, r_N2O_case2 vs. r_N2O_AIT: r = 0.48, p < 0.001; Figure 6). For the other two treatments amended with Glu_0.5 and Glu_1.0, although Bland–Altman analysis revealed significant mean biases in the agreement assessment between MAP- and AIT-estimated r_N2O values, both the 95% limits of agreement and Pearson correlation analysis demonstrated reasonable consistency between the r_N2O values estimated by these two methods. Previous studies often limited comparisons between the δ¹⁵N^SP/δ¹⁸O MAP and AIT to simple differences in r_N2O [5,38]. Few, like the present study, have used this statistical analysis to assess the agreement between the r_N2O derived from both methods.

4.3. Implications and Limitations of the Glucose Amendment Experiment

Based on the δ¹⁵N^SP/δ¹⁸O MAP, we found that carbon availability modulated microbial pathways and consequently influenced N₂O production and consumption. Our results highlighted that fungal contribution was strengthened with increasing glucose concentrations, further resulting in increased emissions of the greenhouse gas N₂O. Thus, we suggest that fungal-derived N₂O constitutes a significant source when assessing N₂O dynamics in vegetable cropping systems, particularly in regions with high N and C inputs. These findings provide better insight into field assessments of N₂O flux in vegetable cropping systems with high fertilizer (N and C) and water inputs. The commonly observed N₂O dynamics may involve diverse contributions from biotic (e.g., nitrification, bacterial/fungal denitrification, nitrifier denitrification) and abiotic (chemo-denitrification) pathways [6,8,30], leading to more complex source apportionment/attribution of N₂O. Based on this study’s finding that impaired bacterial denitrification (i.e., enhanced incomplete denitrification) promotes N₂O accumulation, it can be established that rational carbon source management that favor bacterial denitrification can achieve N₂O mitigation in vegetable cropping systems.

5. Conclusions

Collectively, our study demonstrates that available carbon substrates critically regulate denitrification dynamics and N₂O emissions. Increased glucose concentrations intensified microbial activity, driving concurrent increases in CO₂ and N₂O fluxes while promoting fungal denitrification (FD). Crucially, despite FD enhancement, bacterial denitrification (BD) remained the predominant pathway throughout most of the incubation period across varying glucose concentrations. These results establish that carbon availability modulates N₂O production and reduction efficiency by shifting the balance between BD and FD pathways. We thus emphasize the critical importance of carbon management strategies for N₂O mitigation. Future studies should prioritize elucidating how complex carbon sources (e.g., cellulose, lignin, humus) influence pathway partitioning and N₂O sinks to inform targeted strategies.