Mechanisms and Mitigation of Nitrate Vertical Transport in Black Soil Croplands of Northeast China: Evidence from a ¹⁵N-Tracing Study

Abstract

In Northeast China’s degraded croplands, nitrate (NO₃⁻-N) leaching is the dominant pathway for fertilizer-nitrogen (N) loss, which presents an increasing threat to the quality of groundwater. Conservation tillage, defined as no-tillage (NT) and straw retention, is a widely adopted management strategy to maintain cropland fertility in the black soil (BS) regions. At present, however, the impact of shifting from conventional to conservation tillage on the vertical distribution and regulatory mechanisms of NO₃⁻-N derived from applied fertilizer-N (F_NO3) remains poorly understood. Based on a 12-year field experiment, we integrated ¹⁵N-tracing field monitoring with ¹⁵N-paired-labeling incubation to quantify the vertical migration of F_NO3 into deep soil profiles, and specify the dominant processes regulating N retention and supply. Across the tested BS croplands, total NO₃⁻-N production rates (4.06–6.58 mg N kg⁻¹ soil day⁻¹) were faster than their consumption rates (0.36–0.92 mg N kg⁻¹ soil day⁻¹), leading to a net accumulation of NO₃⁻-N, and implying a potential for leaching of NO₃⁻-N, from the perspective of substrate availability. The results of the field ¹⁵N micro-plot experiment also indicated that, by maize maturity in the first growing season, an average of 7.5% of F_NO3 had migrated to the 80–100 cm soil layer. During the following two growing seasons, the maximum accumulation of F_NO3 had shifted downward to 140–160 cm and 180–220 cm, respectively. Such a pattern, particularly in light of the increased extreme precipitation in the studied regions, raises clear concerns about NO₃⁻-N leaching losses. Compared with conventional management, no-tillage with full-rate straw mulching decreased net rates of NO₃⁻-N production from 6.22 to 3.14 mg N kg⁻¹ soil day⁻¹. This reduction resulted from a decline in the gross oxidation of NH₄⁺-N to NO₃⁻-N (from 6.39 to 3.70 mg N kg⁻¹ soil day⁻¹) and an increase in DNRA (from 0.35 to 0.85 mg N kg⁻¹ soil day⁻¹), which collectively delayed the downward transport of F_NO3. Conservation tillage also increased the gross rate of heterotrophic nitrification (from 0.19 to 0.36 mg N kg⁻¹ soil day⁻¹) and its proportion relative to total nitrification (from 2.8% to 8.9%). Despite this shift, autotrophic nitrification remained the dominant process for NO₃⁻-N production in the tested BS croplands, likely due to a pH constraint on heterotrophic nitrification. With the increasingly widespread promotion of conservation tillage for soil fertility improvement, heterotrophic nitrification warrants greater attention, particularly in BS regions where pH < 6.5 and C/N contents are relatively high. Collectively, our findings provide a scientific basis for tailoring tillage practices to maintain sustainable agriculture in Northeast China.

1. Introduction

The black soil (BS) region of Northeast China is a critical granary for national grain production, serving as the cornerstone in safeguarding national food security [1]. Decades of conventional intensive tillage and unsustainable straw removal, however, have degraded these croplands, manifesting in topsoil erosion, nutrient depletion, and soil compaction [2]. Sustained high crop yields, in turn, have become heavily dependent on additional fertilizer-N application [3]. Excessive N inputs further intensify the environmental risks of N losses [4], exacerbating the contradiction of increasing grain production versus improving fertilizer-N use efficiency (NUE) across the degraded croplands [5].

In their study, Wang et al. (2025) [6] found that, across the global maize and wheat cropping systems during 1961–2020, NO₃⁻-N leaching was the main factor responsible for fertilizer-N loss. In the BS croplands of Northeast China, a conservation tillage (CT) strategy, based on no-tillage (NT) and at least 30% straw mulching (termed the ‘Lishu-Mode’), has been widely implemented to alleviate soil degradation [7,8]. Against the backdrop of tillage transition, it is important to clarify the vertical migration and regulatory mechanisms of fertilizer-derived NO₃⁻-N (F_NO3). Such understanding is a critical step towards developing evidence-based N management strategies that promote grain productivity and environmental sustainability in the degraded croplands.

Upon application to soil, fertilizer-N accumulates and is sequentially transported among the different N pools, including soil NH₄⁺-N, NO₃⁻-N, fixed NH₄⁺-N, and organic-N [9]. Ultimately, it is partitioned into three major routes: (i) uptake by the seasonal crop, which determines fertilizer NUE and crop yield [10]; (ii) retention in the soil system as residual fertilizer-N, which exerts a residual N effect on subsequent crops [11]; and (iii) loss of fertilizer-N from soil–crop systems with corresponding environmental effects [12]. Consequently, using ¹⁵N stable isotope tracing initiated in long-term field experimental platforms makes it possible to perform quantitative analyses of the vertical movement of F_NO3 under contrasting tillage management practices [13], which provides a more effective assessment of NO₃⁻-N leaching potential in degraded agroecosystems.

Within the soil–crop–environment system, the allocation of fertilizer-N is primarily regulated by soil N transformation processes and their respective rates [14,15]. Although the net nitrification rate may still provide information on NO₃⁻-N availability and the associated potential risk of leaching loss [16,17], a low net nitrification rate does not necessarily imply slow turnover dynamics of soil NO₃⁻-N, given the fact that rapid gross nitrification coinciding with equally rapid microbial immobilization can also ensure that soil maintains a low NO₃⁻-N concentration [18]. As such, solely considering net N transformation rates is inadequate to identify the specific processes that dominate N supply and conservation in soil [19]. Gross N transformation rates (GNTR), in contrast, offer a more mechanistic view by simultaneously quantifying production and consumption rates of individual N processes [20]. Shifting the analytical emphasis from net changes to GNTR thus provides a process-based insight into the underlying mechanisms that drive efficient N supply and retention in agricultural soils [21,22].

The results of recent studies have demonstrated that soil organic carbon, total nitrogen, microbial diversity, and composition reach relatively stable states after 8–10 years of continuous maize straw mulching [23,24]. Based on a 12-year field experiment, we aimed to accomplish the following three linked objectives: (1) a ¹⁵N-tracing micro-plot experiment was performed over three maize growing seasons to accurately quantify the vertical migration patterns of F_NO3 in deep soil profiles under different tillage management strategies; (2) we combined ¹⁵N paired-labeling incubation with an MCMC-based numerical analysis model to elucidate the turnover dynamics of GNTR during the tillage transition; and (3) we investigated the key processes regulating the retention of nitrate-N and mitigated its leaching potential from the perspective of substrate availability. These findings ultimately provide scientific evidence highlighting NO₃⁻-N leaching potential and its regulatory mechanisms in the BC croplands of Northeast China.

2. Materials and Methods

2.1. Site Description

A long-term field experiment was conducted at the Lishu Conservation Tillage Research and Development Base (43°19′ N, 124°14′ E) in Jilin Province. The site is located in a typical temperate semi-humid continental monsoon climate zone. The average amount of rainfall is 614 mm per year, with roughly 75% of the average rainfall distributed over the period from June to September. The average annual air temperature is 6.9 °C, and the lowest and highest monthly mean temperatures are recorded in January (–13.5 °C) and July (23.7 °C), respectively. The soil is a Mollisol with a loamy clay texture [25]. Baseline physicochemical properties in the 0–20 cm soil layer in 2007 are shown in

Table 1. Soil physicochemical properties in the 0–20 cm layer of 2007.

Physical Properties			Chemical Properties
Soil texture (%)	Sand	28.5	Available nutrient (mg kg−1)	Alkeline N	90.1
	Silt	38.6		Available P	6.9
	Clay	32.9		Available K	143.6
Clay mineral composition (<2 µm, %)	Chlorite	30.0	Total nutrient (g kg−1)	Soil organic C Total N Total P Total K	11.3 1.2 0.38 24.3
	Montmorillonite	24.2
	Illite	14.5
	Vermiculite	2.7
	Kaolinite	23.3
	Quartz	5.0
	Feldspar	0.3

Note: Sand: 2–0.05 mm; Silt: 0.05–0.002 mm; Clay: <0.002 mm.

.

2.2. Long-Term Field Experimental Platform

The long-term field experiment followed a randomized block design with four treatments: (1) traditional ridge tillage (RT), and three CT managements—(2) NT with maize straw removed (NT), (3) NT with reduced straw mulching (2500 kg ha⁻¹, NT + RS), and (4) NT with full-rate straw mulching (7500 kg ha⁻¹, NT + FS). Each of the four treatments was replicated four times in the individual main plot (8.7 m × 30 m).

A continuous maize monoculture system was adopted throughout the studied site. In the RT treatment, soil was plowed to a depth of 25–30 cm in autumn after maize harvest annually, and all aboveground maize debris was removed. For CT treatments, sowing and fertilization were simultaneously conducted using a 2BMZF-4 no-till planter (Jilin Kangda Agricultural Machinery Co., Ltd., Jilin, China). For NT + RS and NT + FS plots, maize straw was applied manually at the designated rates. Spring maize cultivar ‘Heyu 9’ was sown at 60,000 plants ha⁻¹. Basal fertilizers were applied at 10−15 cm soil depth, with application rates of 240 kg N ha⁻¹, 110 kg P₂O₅ ha⁻¹ and 110 kg K₂O ha⁻¹.

2.3. ¹⁵N-Tracing Field Micro-Plot Experiment

In May of 2016, after 9 years of continuous application of CT management, 2 m × 2 m micro-plots bounded by PVC boards were established in each of the 16 main plots (4 treatments × 4 replications). Considering the main distribution zone of the maize root in the studied site and minimizing soil disturbance to the NT agroecosystem, the PVC boards were inserted 60 cm below the ground and extended 10 cm above the surface. For each ¹⁵N micro-plot, fertilizers were applied at the same rate as their respective main plots. The sole difference was the use of urea labeled with ¹⁵N abundance of 10% at spring sowing in the first growing season, whereas unlabeled urea was used in the subsequent growing seasons.

Following three years of continuous ¹⁵N-tracing micro-plot observation, soil samples were collected annually after maize maturity using a 2.5 cm diameter soil auger. Composite samples (from three points in each micro-plot) were taken at 20 cm intervals within a soil profile ranging from 0 to 100 cm (five soil layers on 14 October 2016) and from 0 to 300 cm (fifteen soil layers on 13 October 2017 and 10 October 2018). The soil samples (16 main plots × 5 or 15 soil layers × 3 replications) were analyzed to determine NO₃⁻-N concentrations and their respective ¹⁵N abundance across the profile, to thus more effectively quantify the vertical distribution and leaching potential of F_NO3 under the different tillage regimes.

2.4. ¹⁵N Paired-Labeling Incubation Experiment

As long-term application of CT practices improved soil fertility predominantly in the near-surface soil layer [26], topsoil (0–10 cm, 6 points, S-shaped) from the 16 main field plots was collected in October 2018 after maize harvest for the determination of GNTR. Fresh samples were promptly transported to the laboratory, where visible gravel, plant roots, and debris were removed. After sieving (2 mm) and homogenization, the soil was split into two parts: one subsample was used to determine soil water holding capacity (WHC) and then stored at 4 °C for the following ¹⁵N-paired labeling incubation experiment, and the other subsample was air-dried for the determination of pH, SOC, and TN content.

For the ¹⁵N-paired labeling incubation experiment, homogenized fresh soil (equivalent to 30 g of dry soil weight) was placed into 250 mL Erlenmeyer flasks, with 24 replicates for each treatment (2 types of ¹⁵N labeling × 3 replicates × 4 sampling times). After 1 day of pre-incubation at 25 °C, labeling solutions, i.e., either ¹⁵NH₄NO₃ (4.85 atom% ¹⁵N excess) or NH₄¹⁵NO₃ (4.89 atom% ¹⁵N excess), were applied dropwise to ensure uniform distribution at a rate of 50 mg NH₄⁺-N or NO₃⁻-N kg⁻¹ soil. Soil moisture was adjusted to 60% of WHC using deionized water. Each flask was sealed with perforated plastic film to maintain the aerobic conditions, and incubation was continued at 25 °C. Three random flasks from each labeling treatment were destructively sampled at 0.5, 12, 24, and 48 h after labeling to measure the concentrations and ¹⁵N abundance of NH₄⁺-N and NO₃⁻-N, respectively.

The measured concentrations and ¹⁵N atom% excess values of NH₄⁺-N and NO₃⁻-N were incorporated into an MCMC-based numerical model of N transformation processes [27]. The model iteratively adjusted the kinetic parameters of individual N transformation pathways to fit the simulated curves to the observed incubation data. The model ultimately output the following three paired GNTR regarding the turnover dynamics of NO₃⁻-N: organic-N mineralization (M_Norg) and NH₄⁺-N immobilization (I_NH4); oxidation of NH₄⁺-N to NO₃⁻-N (O_NH4) and dissimilatory NO₃⁻-N reduction to NH₄⁺-N (DNRA); and oxidation of organic-N to NO₃⁻-N (O_Nrec) and NO₃⁻-N immobilization (I_NO3). It should be noted that the model specifically parameterized DNRA and I_NO3 as the primary biological consumption pathway for NO₃⁻-N. In comparison, other potential NO₃⁻-N loss pathways, such as denitrification (which would require the measurements of ¹⁵N gas flux), were not directly quantified in this incubation experiment. For further details on the ¹⁵N-paired labeling incubation experiment, refer to the works of Xie et al. (2018) [28] and Zhang et al. (2023) [29].

2.5. Analysis of Basic Soil Properties

Soil physicochemical properties were determined following the Soil Agro-Chemical Analysis Procedures [30]. In detail, soil pH was measured with a DMP 2 mV/pH meter. Soil WHC was determined using the oven-drying method. SOC was quantified via potassium dichromate oxidation. Soil TN was quantified using an elemental analyzer (Elementar Vario MACRO cube, Langenselbold, Germany). NH₄⁺-N and NO₃⁻-N were extracted with 2M KCl, and their concentrations were determined on a continuous-flow chemical analyzer (SmartChem 200, AMS, Guidonia Montecelio, Italy). ¹⁵N abundance was measured using an isotope-ratio mass spectrometer (DELTAplus XP, Thermo Finnigan, Waltham, USA).

2.6. Data Calculations

The content of F_NO3 and its proportion in the total applied urea-N (P_NO3) were quantified as follows [31]:

$${\mathrm{S}}_{\mathrm{N}\mathrm{O}3} = \mathrm{C}\times \mathrm{H}\times \mathsf{\rho }\times {10}^{-1}$$

(1)

$${\mathrm{F}}_{\mathrm{N}\mathrm{O}3}={\mathrm{S}}_{\mathrm{N}\mathrm{O}3}\times {\displaystyle \frac{\mathrm{a}-\mathrm{c}}{\mathrm{b}-\mathrm{c}}}$$

(2)

$${\mathrm{P}}_{\mathrm{N}\mathrm{O}3}={\displaystyle \frac{{\mathrm{F}}_{\mathrm{N}\mathrm{O}3}}{\mathrm{F}\mathrm{N}}}\times 100\%$$

(3)

where S_NO3 is the stock of NO₃⁻-N in the soil profile (kg N ha⁻¹), $\mathrm{C}$ is the concentration of NO₃⁻-N in a given soil layer (mg N kg⁻¹), H is the thickness of a given soil layer (cm), $\mathsf{\rho }$ is the bulk density of a given soil layer (g cm⁻³), and 10⁻¹ is a unit conversion factor. a is the ¹⁵N abundance of NO₃⁻-N in a given soil layer; b is the ¹⁵N abundance of applied urea-N (10%); c is the natural ¹⁵N abundance (0.3663%); and $\mathrm{F}\mathrm{N}$ is the amount of urea-N applied (240 kg N ha⁻¹).

The associated GNTR were calculated as follows [19,20]:

Gross NO₃⁻-N production rate (G_NO3) = O_NH4 + O_Nrec

(4)

Gross NO₃⁻-N consumption rate (C_NO3) = DNRA+ I_NO3

(5)

Net NO₃⁻-N production rate (N_NO3) = G_NO3 − C_NO3

(6)

Data are expressed as the mean ± standard deviation using Microsoft Excel 2021. ANOVA followed by the LSD test was conducted to evaluate the treatment effects on the proportion of F_NO3, soil basic physicochemical properties, and GNTR. Correlation analysis was performed to examine the relationships between these variables. Differences were considered statistically significant at p < 0.05. All the statistical analyses were conducted in R 4.2.1 [32], and figures were prepared using OriginPro 2022.

3. Results

3.1. Distribution of F_NO3 in the Soil Profile

During the first growing season after urea-N application, the results of the ¹⁵N-tracing field micro-plot experiment indicated that, roughly 8.9–10.5% (average 9.7%) of the applied ¹⁵N-urea persisted as NO₃⁻-N within the 0–100 cm soil profile (Figure 1A). Within this fraction, more than 74.7% of F_NO3 had already accumulated in the 80–100 cm soil layer (Figure 2A).

By the second growing season, after further uptake by the crop, approximately 6.5–8.6% (average 7.1%) of the initial ¹⁵N-urea remained as NO₃⁻-N across the 0–300 cm profile (Figure 1B). The peak accumulation of F_NO3 appeared to shift downward, concentrating in the 120–160 cm soil layer (Figure 2B).

For the third growing season, approximately 4.7–7.6% (average 5.8%) of the applied ¹⁵N-urea was still present as NO₃⁻-N in the 0–300 cm soil profile (Figure 1C). Correspondingly, the F_NO3 peak was observed to migrate deeper to the depth of 180–220 cm (Figure 2C).

Across the three consecutive growing seasons, compared with the RT treatment, no-tillage with full-rate straw mulching (NT + FS) significantly decreased the proportion of F_NO3 relative to the total applied ¹⁵N-urea in the soil profile (Figure 1), and also appeared to effectively delay the downward migration and accumulation of F_NO3 in the deeper soil layer (Figure 2).

3.2. Soil Physicochemical Properties

The soil pH in the studied BS croplands ranged between 6.7 and 7.5. Compared with the RT treatment, both reduced (NT + RS) and full-rate (NT + FS) straw mulching under no-tillage management significantly decreased soil pH by 0.3 and 0.6 units, respectively (Figure 3A). Across all the tillage regimes, the soil WHC, SOC, and TN contents varied, spanning 61.8–69.4%, 11.67–15.60 g kg⁻¹, and 1.28–1.68 g kg⁻¹, respectively. In contrast to the soil pH response, no-tillage with straw mulching consistently increased the WHC, SOC, and TN contents compared with RT. These increases were significant for the full-rate straw mulching treatment (NT + FS), which enhanced WHC by 7.6% (Figure 3B), SOC by 25.1% (Figure 3C), and TN by 15.8% (Figure 3D), respectively.

3.3. Gross N Transformation Rates

The results of the ¹⁵N-paired-labeling incubation experiment showed that, throughout the incubation process, the NH₄⁺-N concentrations decreased gradually across all the tillage management strategies (Figure 4A); in comparison, the NO₃⁻-N concentrations increased correspondingly (Figure 4B), denoting net nitrification in the studied BS croplands. After 100,000 iterations, the MCMC-based numerical model accurately simulated the dynamics of inorganic-N concentrations (Figure 4A,B) and their corresponding ¹⁵N atom% excess values (Figure 4C–F), as the simulated curves consistently fluctuated within the variability of the measured data (mean ± standard deviation). Quantitative fit diagnostics further indicated high R² values (typically > 0.95) for all the treatments, confirming the robust performance and reliability of the MCMC-based model in simulating the turnover dynamics of the soil N cycle. The final output of the paired GNTR for each tillage management strategy is summarized in Figure 5.

G_NO3 in the studied BS croplands ranged from 4.06 to 6.58 mg N kg⁻¹ soil day⁻¹ (

Table 2. Dynamics of the gross NO₃⁻-N production and consumption rates under different tillage management strategies.

Parameters	RT	NT	NT + RS	NT + FS
GNO3	6.58 ± 0.24 a	5.99 ± 0.14 b	5.31 ± 0.21 c	4.06 ± 0.16 d
ONH4/GNO3 (%)	97.22 ± 0.84 a	96.03 ± 0.71 ab	95.22 ± 0.22 b	91.13 ± 0.86 c
ONrec/GNO3 (%)	2.78 ± 0.84 c	3.97 ± 0.71 bc	4.78 ± 0.22 b	8.87 ± 0.86 a
CNO3	0.36 ± 0.05 c	0.48 ± 0.15 bc	0.69 ± 0.16 ab	0.92 ± 0.16 a
DNRA/CNO3 (%)	96.73 ± 0.76 a	96.04 ± 0.17 a	94.97 ± 2.07 a	92.61 ± 0.87 b
INO3/CNO3 (%)	3.27 ± 0.76 b	3.96 ± 0.17 b	5.03 ± 2.07 b	7.39 ± 0.87 a
NNO3	6.22 ± 0.21 a	5.51 ± 0.03 b	4.62 ± 0.32 c	3.14 ± 0.09 d
ONH4/INH4 (N/I)	1.85 ± 0.18 a	1.50 ± 0.05 b	1.30 ± 0.18 b	0.77 ± 0.04 c

Note: Different lowercase letters in the same row indicate significant differences among treatments at p < 0.05.

). Autotrophic nitrification (O_NH4) contributed over 90% of the total G_NO3, whereas heterotrophic nitrification (O_Nrec/G_NO3) accounted for less than 10%. Compared with the RT treatment, all the CT managements (NT, NT + RS, and NT + FS) significantly reduced the gross rate of O_NH4 (Figure 5), and thus, the total G_NO3 decreased by 8.8%, 19.3%, and 38.3%, respectively.

C_NO3 in the studied BS croplands varied from 0.36 to 0.92 mg N kg⁻¹ soil day⁻¹ (Table 2), among which DNRA accounted for more than 93% of the total NO₃⁻-N consumption pathways parameterized in our model. Both NT + RS and NT + FS significantly increased DNRA relative to RT (Figure 5), and consequently, enhanced the total C_NO3 by 92.0% and 156.4%, respectively.

Across the treatments, gross NO₃⁻-N production (G_NO3) was consistently higher than the corresponding consumption (C_NO3), which resulted in net NO₃⁻-N production (N_NO3 ranging between 3.14 and 6.22 mg N kg⁻¹ soil day⁻¹). CT management decreased G_NO3 while increasing C_NO3 and accordingly led to respective declines in N_NO3 of 11.4%, 25.7%, and 49.6% for NT, NT + RS, and NT + FS compared to RT (Table 2). Ratios of O_NH4 to I_NH4 (N/I) in the studied BS ranged from 0.77 to 1.85, which decreased by 18.6%, 29.9%, and 58.5% under the NT, NT + RS, and NT + FS treatments, respectively (Table 2).

3.4. Correlation Analysis

Correlations between soil physicochemical properties and GNTR are provided in Figure 6. In detail, O_NH4 correlated positively with soil pH, but negatively with both SOC content and the gross rate of I_NH4. In contrast, O_Nrec exhibited a significant positive correlation with the WHC, SOC, and TN content, but was inversely related to soil pH. Both the gross rate of I_NO3 and DNRA were positively correlated with the SOC content across all the tillage management strategies (p < 0.05, n = 16).

4. Discussion

4.1. Leaching Potential

The fates of fertilizer-N in soil–crop–environment systems are largely controlled by soil N transformation processes and their respective rates [14,15]. As G_NO3 was significantly higher than C_NO3 in all tillage management strategies, net production of NO₃⁻-N was consistent in the studied BS croplands (Table 2). Similar results were also recorded in our ¹⁵N paired-labeling incubation experiment, during which the NH₄⁺-N concentration decreased gradually, whereas the NO₃⁻-N concentration increased progressively (Figure 4A,B). In a recent study by Manono et al. (2026) [33], the authors revealed that nitrogen leaching from ecosystems is driven by a complex interplay of environmental factors, including soil properties (e.g., availability of NO₃⁻-N and microbial activity), topography, and hydrology (e.g., preferential flow through macropores). These lines of evidence collectively suggest that, from the perspective of substrate availability, F_NO3 is prone to leaching losses in the studied spring maize agroecosystems. The results of the field ¹⁵N-micro-plot experiment further indicated that, after three years of continuous tracing, F_NO3 was observed to migrate into the 180–220 cm soil layer (Figure 2C). Such a rapid vertical redistribution, particularly under the increasing frequency of extreme precipitation events recorded for Northeast China [34], implies that NO₃⁻-N leaching potential and its associated threat to regional groundwater pollution cannot be overlooked.

4.2. Mitigation Mechanisms

Soil N cycling is essentially a microbe-driven process, which mediates the turnover dynamics between inorganic and organic-N pools [35]. Shifting tillage practices reshapes the soil N cycle through two interlinked mechanisms: (i) a substrate effect, by which changing TN content directly modifies the substrate available for soil N cycling [36,37], and (ii) a microbe effect, whereby shifts in soil key environmental conditions—such as pH, WHC, availability of SOC, and TN content—affect microbial activity and indirectly regulate the rates and products of the soil N transformation processes [38,39]. Our findings further support this synergistic view, as verified by the significant correlations between the paired GNTR and soil physicochemical properties (Figure 6). Consequently, elucidating the turnover dynamics of individual NO₃⁻-N production and consumption processes under different tillage regimes, based on the framework of ‘substrate–microbe interplay’ [15], is essential to identify the specific mechanisms that regulate both the efficient retention and supply of NO₃⁻-N in the studied BS croplands.

Reduced total NO₃⁻-N production in CT agroecosystems: In the tested BS croplands, O_NH4 accounted for more than 90% of the total G_NO3 (Table 2). With soil pH ranging from 6.7 to 7.5 (Figure 3A), ammonia-oxidizing bacteria (AOB) represent the key functional microbes driving nitrification in agricultural soils [40,41]. These results collectively reveal that NO₃⁻-N production in the tested BS croplands is primarily derived from AOB-dominated autotrophic nitrification, which was in accordance with the results of previous studies conducted by Zhang et al. (2013) [42] and Elrys et al. (2023) [22]. Relative to RT, NT significantly reduced the gross rate of M_Norg (Figure 5). This suppression likely results from the lower soil disturbance and higher soil aggregation in the CT agroecosystems [43], which physically protect organo-mineral complexes and limit microbial access to organic-N substrates [44]—a strategy more effective in clay-rich soils [45]. Straw mulching also improves soil micro-environmental conditions (such as moisture and C/N availability), thereby stimulating microbial NH₄⁺-N immobilization (I_NH4) [46]. Consequently, a key factor governing the low total G_NO3 under CT agroecosystems is the limited supply of substrate (NH₄⁺-N) for O_NH4. CT management strategies, particularly with full-rate straw mulching, also considerably reduced soil pH (Figure 3A). This acidification is likely to have resulted from the leaching of base cations (such as K⁺, Ca²⁺, Na⁺, and Mg²⁺), accelerated by the organic acids released during straw decomposition [47], which can subsequently suppress the abundance and activity of AOB [39]. Combining this microbial effect with the substrate limitation mentioned above, no-tillage and straw mulching dramatically decreased the ratio of O_NH4/I_NH4 (Table 2). This shift, in turn, reflects a more efficient N conservation cycle established in the CT management strategies [48,49], and ultimately reduced total NO₃⁻-N production in CT agroecosystems.

Enhanced total NO₃⁻-N consumption in CT agroecosystems: I_NO3 and DNRA were identified as the two major biological consumption pathways for soil nitrate. Across all tillage management strategies, the gross rate of I_NO3 was negligible relative to I_NH4 (Figure 5), which could be attributed to the high energy cost required for I_NO3 [50]. As a result, among the NO₃⁻-N consumption pathways directly quantified using our ¹⁵N model, DNRA accounted for over 93% of the total C_NO3 in the tested BS croplands (Table 2). This finding is consistent with the previous study, in which the results indicated that, in neutral to alkaline agricultural soils, DNRA serves as the predominant process for NO₃⁻-N consumption [51]. Relative to RT, no-tillage and straw mulching significantly stimulate the DNRA process. This improvement can mainly be attributed to the increased availability of labile SOC and TN substrates [52]. By sequestering NO₃⁻-N into the plant- microbe-available NH₄⁺-N form, CT managements thus substantially increased total NO₃⁻-N consumption, and thereby prevented its potential loss through leaching [53].

4.3. Management Implications

Relative to RT, CT management strategies significantly enhanced both the gross rate of heterotrophic nitrification (O_Nrec, Figure 5) and its proportion in total nitrification (O_Nrec/G_NO3, Table 2). Two factors likely explain this stimulation: (i) long-term CT practices increase the stock of soil organic-N [54,55], maintaining a constant substrate supply for O_Nrec, and (ii) the improved soil micro-environment conditions under CT managements likely favor the abundance and activity of fungi [56], which are recognized as key drivers of heterotrophic nitrification in soil [57]. It is noteworthy that the observed O_Nrec/G_NO3 ratio for the tested BS croplands fell within the range of 3% to 9%, which is substantially lower than the global average ratio of 32%, reported by Zhang et al. (2023) [58] from 491 observations for agroecosystems. This inconsistency could be related to the pH threshold of O_Nrec, which is generally considered negligible in soils with pH > 6.5 [59].

Despite its relatively low contribution, no-tillage with full-rate straw mulching nevertheless doubled the ratio of O_Nrec/G_NO3 relative to conventional management. As 54.8% of croplands in the tested BS regions are characterized by a pH lower than 6.5 [60], and regional variations in SOC and TN contents are significant across Northeast China [61], widespread application of CT management—particularly in areas with lower soil pH (<6.5) and higher C/N contents—could substantially stimulate O_Nrec. Under these conditions, heterotrophic nitrification can contribute more substantially to total nitrification, which should not be overlooked. In order to more effectively quantify the spatial variability of O_Nrec and its impact on soil N cycling, future studies should conduct experiments at different soil pH gradients and C/N contents within BS croplands. Such efforts will refine our theoretical insight into soil NO₃⁻-N transformation dynamics, and thus provide tailored strategies for the sustainable development of regional agriculture.

4.4. Experimental Limitations

In the ¹⁵N-tracing field micro-plot experiment, the PVC boards were simply inserted to a depth of 60 cm, making it impossible to exclude the potential for lateral migration of F_NO3 throughout the deep soil profile. Future studies could address this limitation by simultaneously characterizing both vertical and lateral distributions of F_NO3 in response to different management practices, thereby providing a more precise assessment of the distribution of F_NO3 in agroecosystems.

Additionally, in this study, we focused exclusively on the availability of substrate (nitrate) as a single factor while neglecting the role of hydrological characteristics in the vertical migration of F_NO3 within the soil profile. As the experimental platform used in this study was based on a long-term field trial, future research could install lysimeters or similar devices to quantitatively collect leachate, thereby enabling a more robust evaluation of F_NO3 leaching losses under different management practices.

In the ¹⁵N-paired-labeling incubation experiment, all the GNTR measurements were conducted on near-surface soil samples based on the assumption that surface processes largely govern the nitrate supply entering the transport domain (source control). Future studies investigating GNTR at different soil depths will enhance our understanding of N turnover dynamics throughout the entire soil profile and provide a better assessment of the contribution of subsoil processes to overall nitrate retention or loss potential in agroecosystems.

5. Conclusions

In the black soil croplands of Northeast China, F_NO3 exhibited a rapid vertical translocation, which was observed to migrate into the 180–220 cm soil layer after three years of continuous observation. As the extreme precipitation has become increasingly frequent in the studied regions, this pattern suggests a potential for leaching of F_NO3. Relative to RT, no-tillage with full-rate straw mulching substantially reduces the deep distribution potential and thus the leaching potential of F_NO3 through two synergistic mechanisms: (i) suppression of autotrophic nitrification to reduce total NO₃⁻-N production, and (ii) enhancement of DNRA to increase total NO₃⁻-N consumption, which significantly decreased net NO₃⁻-N production, and thereby delayed the downward transport of F_NO3 into the deeper soil layer. It should also be noted that, while autotrophic nitrification remains the dominant pathway for NO₃⁻-N production in the tested black soils, CT practices significantly stimulate both the turnover dynamics of heterotrophic nitrification and its proportion in total nitrification. In light of the widespread adoption of conservation tillage in recent management guidelines and the evident high leaching potential in the studied regions, critical consideration should be given to the role of heterotrophic nitrification in soil N cycling, particularly in black soil regions with a pH below 6.5 and relatively high SOC and TN contents. These findings offer a scientific basis for designing site-specific tillage strategies for sustainable agriculture in Northeast China.