Alteration of Nitrogen Fertilizer Forms Optimizes Nitrogen Balance in Drip-Irrigated Winter Wheat Systems of Northern China by Reducing Gaseous Nitrogen Losses

Abstract

Winter wheat covers approximately 2.21 × 10⁸ ha globally, making it the most widely cultivated cereal crop in the world. In recent years, integrated water and fertilizer management has significantly improved winter wheat yield and nitrogen use efficiency; however, quantitative assessments of nitrogen cycling under different fertilizer forms in such high-yield systems remain limited. From 2022 to 2024, a two-year field experiment was conducted in drip-irrigated winter wheat fields in northern China. Four nitrogen fertilizer forms were applied: nitrate nitrogen fertilizer (NON), ammonium nitrogen fertilizer (NHN), amide nitrogen fertilizer (CON), and urea ammonium nitrate fertilizer (UAN), along with an unfertilized control (CK). Compared with NON, NHN, and CON, UAN reduced cumulative N₂O emissions by 10.40–15.64% and NH₃ volatilization by 2.04–9.33% (p < 0.05). It also increased the leaf area index and biomass accumulation at maturity, as well as grain yield (3.70–10.28%), nitrogen harvest index (4.58–12.88%), and nitrogen use efficiency (12.14–39.25%) (p < 0.05). Furthermore, UAN significantly decreased the net nitrogen surplus (24.18–45.70%) and nitrogen balance values (25.64–55.82%) (p < 0.05). Correlation analysis indicated that the reduction in nitrogen balance was primarily attributed to lower N₂O emissions and improved nitrogen use efficiency (p < 0.05). In conclusion, the application of urea ammonium nitrate under integrated water–fertilizer management achieved higher yield, greater efficiency, and environmentally sustainable production in drip-irrigated winter wheat systems in northern China.

1. Introduction

Northern China is the major winter wheat production region and plays a vital role in ensuring national food security. However, water scarcity and low nitrogen (N) use efficiency have long constrained high yields and sustainable development in this area. In recent years, water–fertilizer integration technology has been increasingly adopted in winter wheat production across northern China [1]. By enabling precise irrigation and fertilization, this approach synchronizes water and nitrogen supply, thereby improving water use efficiency (WUE) and nitrogen use efficiency (NUE). Studies have shown that, compared with traditional flood irrigation and broadcast fertilization, water–fertilizer integration with proper N management can increase winter wheat yield by 10.4–12.6%, with maximum yields reaching 12,080 kg ha⁻¹ [2]. Multi-site trials across northern China have consistently reported yields exceeding 10,500 kg ha⁻¹, representing a super-high-yield level [2,3,4]. Yield improvements under this system are mainly attributed to increases in spike number and thousand-grain weight at maturity, enhanced post-anthesis biomass accumulation, and greater flag leaf photosynthetic activity [4]. Moreover, water–fertilizer integration improves the spatial and temporal coordination of roots, nutrients, and water, significantly enhancing WUE and NUE while reducing deep-soil nitrate (NO₃⁻-N) accumulation [5]. Under such systems, different nitrogen fertilizer forms (urea, ammonium nitrate, and ammonium sulfate) show distinct effects on crop yield, WUE, NUE, and nitrogen losses [6]. Using the ¹⁵N labeling technique, studies on drip-fertigated winter wheat revealed that nitrate-N exhibited significantly higher NUE than ammonium-N and urea [7]. Modeling analyses further indicate that nationwide adoption of water–fertilizer integration in China could increase grain production by approximately 12% on average, raise WUE and NUE by 26.4% and 34.3%, respectively, and effectively mitigate N leaching [8]. Therefore, water–fertilizer integration provides a crucial technical foundation for overcoming water and nutrient constraints in the arid and semi-arid regions of northern China, enabling high-yield, efficient, and environmentally sustainable winter wheat production.

N is a key element for plant growth, and the availability of soil N exerts a significant influence on crop development [9]. Fertilizer application can improve crop growth and ultimately increase yield. Since the Green Revolution, synthetic fertilizers have been widely used to meet crop N demands. At present, China applies approximately 3 × 10¹¹ kg of N fertilizer (pure N) annually, accounting for one-third of global consumption [10]. Although further increases in N input may help to meet future food demand, the resource-intensive production pattern in agriculture has led to low NUE and considerable N losses (40–50%), particularly in the arid and semi-arid regions of northern China [11]. Studies have shown that applying all N fertilizer as a basal dose results in up to 80% of N remaining in the soil as inorganic forms that cannot be taken up by winter wheat, thereby exacerbating environmental risks [12]. Under water–fertilizer integration, inappropriate irrigation and fertilization practices can aggravate N losses from farmland, with NO₃⁻-N leaching polluting aquatic systems and gaseous losses of nitrous oxide (N₂O) and ammonia (NH₃) increasing environmental burdens [13]. Moreover, when applied as single N forms, fertilizers often exhibit spatial and temporal mismatches between soil N availability and crop root uptake dynamics, leading to inefficiencies where part of the N is either lost as gaseous emissions or remains in the soil [14]. Therefore, within the framework of high-yield water–fertilizer integration systems for winter wheat, optimizing fertilizer forms may represent an effective strategy to further enhance yield and NUE while mitigating N losses.

To improve NUE in wheat, extensive studies have been conducted on N fertilizer types, application rates, and their combination with N inhibitors. It has been reported that ammonium-based N fertilizers promote chlorophyll formation during the seedling stage, thereby enhancing wheat growth [15]. However, some studies suggest that N fertilizer forms have no significant effect on wheat yield [16]. This inconsistency may be attributed to differences in soil type and inherent fertility, as well as the distinct transformation processes of different N fertilizers in soil, which ultimately affect N uptake by wheat [14]. Regardless of whether crops prefer ammonium or nitrate, mixed ammonium–nitrate fertilizers generally lead to higher yields compared with sole ammonium- or nitrate-based fertilizers, while leaving less residual N in the soil. This may be because crops exhibit different N preferences across growth stages, thereby making better use of soil-available N [17]. Research on N application rates has demonstrated that appropriate planting density and optimal N input are beneficial for maintaining high yield and improving NUE [18], with modeling studies further supporting these findings [19]. In addition, previous studies have investigated the effects of N management strategies on wheat yield, NUE, and soil N accumulation. Rational N management can supply sufficient N during key growth stages, thereby improving wheat N uptake and NUE while reducing residual soil N accumulation [20]. Moreover, nitrification inhibitors have been shown to improve wheat yield and NUE by maintaining higher ammonium (NH₄⁺-N) concentrations and NO₃⁻-N concentrations in soil, thus enhancing NUE [21]. Despite these advances, most studies have been based on conventional urea or other solid N fertilizers. With the continuous advancement of fertilizer reduction and efficiency enhancement technologies in northern China, and given the pronounced advantages of water-soluble and liquid nitrogen fertilizers in water–fertilizer integration systems, liquid urea ammonium nitrate (UAN) fertilizer demonstrates great potential for widespread application.

Northern China is the major production region of winter wheat and plays a crucial role in ensuring national food security. To overcome the constraints of limited water resources and low NUE on wheat yield, recent studies have mainly focused on the effects of water–fertilizer integration on yield improvement and efficiency enhancement. However, under high-yield cultivation systems, quantitative studies on the impacts of different N fertilizer forms on soil mineral N residuals and gaseous N emissions in winter wheat fields remain limited. Therefore, a two-year field experiment was conducted in high-yield demonstration fields of the Loess Plateau with the objectives of: (1) quantifying the effects of different N fertilizer forms on soil mineral N residuals and gaseous N emissions during the winter wheat growing season; (2) determining their impacts on crop growth, yield, and nitrogen use efficiency; and (3) establishing a suitable fertilization strategy for high-yield drip-irrigated winter wheat systems in northern China to achieve high yield, high efficiency, and sustainable production. This study hypothesizes that, compared with single-form N fertilizers, compound N fertilizers could increase winter wheat yield while optimizing nitrogen balance in agroecosystems.

2. Materials and Methods

2.1. Experimental Site

The field experiment was conducted over two consecutive years (2022–2024) at the Shenfeng Village experimental station of Shanxi Agricultural University (37°42′ N, 112°53′ E). The site is located in the northeastern part of the Jinzhong Basin on the Loess Plateau, China, at an altitude of 791 m. The region is characterized by a warm temperate continental climate typical of arid and semi-arid zones, with a frost-free period of 159–167 days, an annual mean temperature of 11 °C, annual sunshine duration of 2500–2600 h, and an average annual precipitation of approximately 458 mm. Before the experiment, the soil had the following properties: soil organic matter 12.57 g kg⁻¹, alkali-hydrolysable N 1.01 g kg⁻¹, available P 33.73 mg kg⁻¹, and available K 126.33 mg kg⁻¹. Precipitation and mean daily temperature during the experimental period (2022–2024) are presented in Figure 1 (data collected from an on-site meteorological station). The region follows a single-cropping system with winter wheat grown annually, and the experimental field had been continuously cultivated with winter wheat prior to the initiation of this study.

2.2. Field Management and Experimental Design

The experiment followed a single-factor randomized block design with five treatments: nitrate N fertilizer (NON, NaNO₃, total N 16%), ammonium N fertilizer (NHN, (NH₄)₂SO₄, total N 21%), amide N fertilizer (CON, urea, CO(NH₂)₂, total N 46%), urea ammonium nitrate fertilizer (UAN, liquid, total N 32%, including nitrate N 8%, ammonium N 8%, and amide N 16%), and a control without fertilizer (CK). Each treatment had three replicates, giving 15 plots of 50 m² each (5 m × 10 m). The winter wheat cultivar (Triticum aestivum L.) ‘Jintai 182’ was sown on 5 October 2022 and 25 September 2023 and harvested on 28 June 2023 and 22 June 2024. The sowing density was 1.8 × 10⁶ plants ha⁻¹ with wide-drill broadcasting. Straw residues were incorporated into the 10–15 cm soil layer, seeding depth was 3–5 cm, row spacing was 22–25 cm, and band width was 10–12 cm (Figure 2). Before sowing, 90 kg N ha⁻¹, 150 kg P₂O₅ ha⁻¹, and 90 kg K₂O ha⁻¹ were applied as basal fertilizers. Liquid N fertilizers were diluted and sprayed evenly on the plots. An additional 90 kg N ha⁻¹ was applied at the jointing stage through fertigation, resulting in a total N input of 180 kg ha⁻¹. During the winter wheat growing season, soil moisture-based supplemental irrigation was applied using drip tapes (Figure 1), and the irrigation volume was calculated according to the following formula [22]:

$$M=10\times \gamma \times H\times ({\beta }_i-{\beta }_j)$$

(1)

where $M$ (mm) is the irrigation amount, $\gamma$ (g cm⁻³) is the soil bulk density at 0–40 cm depth, $H$ (cm) is the soil layer depth, ${\beta }_i$ (%) is the target soil moisture content, and ${\beta }_j$ (%) is the soil moisture content before irrigation.

Field capacity was measured using a handheld soil moisture meter (HS2 Hydro Sense II, Campbell, CA, USA). Irrigation was applied at the jointing and anthesis stages when field capacity dropped below 65%, and water was replenished to 80% of field capacity [23]. During the 2022–2024 winter wheat growing seasons, irrigation was applied twice at the jointing stage and once at the anthesis stage, with 37.5 mm of water each time (Figure 1). All other field management practices followed conventional high-yield cultivation practices for winter wheat.

2.3. Measurements and Calculation Methods

2.3.1. Soil NO₃⁻-N and NH₄⁺-N Measurement

Sampling at key growth stages of winter wheat, including the jointing, booting, flowering, and maturity stages, can accurately capture the physiological and biochemical dynamics during these critical periods while establishing the temporal foundation for modeling nitrogen cycling in winter wheat. Therefore, during the winter wheat growing seasons from 2022 to 2024, soil samples were collected at the jointing, booting, anthesis, and maturity stages. In each plot, five sampling points were randomly selected using a 5-cm diameter soil auger, and soil cores were taken at 20-cm intervals from 0 to 100 cm depth. The concentrations of NO₃⁻-N and NH₄⁺-N in the soil were determined using a continuous flow analyzer (Autoanalyzer 3, SEAL, Norderstedt, Germany).

2.3.2. Measurement of N₂O Flux

The N₂O fluxes at the experimental site were measured using the static chamber–gas chromatography method. The static chamber consisted of a sampling box and a base. The sampling box was a polyethylene cube (0.6 m × 0.6 m × 0.6 m) with a layer of thermal insulation material on the surface to minimize temperature fluctuations inside the chamber during sampling. A battery-powered miniature fan was installed in a groove at the top to ensure continuous air circulation. The base was a quadrangular structure (0.6 m × 0.6 m × 0.2 m; Figure 2). After wheat sowing, the base was inserted into the soil at the center of each plot to a depth of approximately 10 cm. The static chamber was sealed by filling the groove with water, and gas samples were collected at 0, 10, 20, and 30 min after chamber closure using 50 mL syringes. Gas sampling was conducted every five days during the experiment, daily in the first week after rainfall and fertilization, every two days in the second week, and every three days in the third week. N₂O concentrations were analyzed using a gas chromatograph (GC-2010 Pro, Shimadzu, Kyoto, Japan), and the N₂O flux was calculated according to the following equation [24]:

$$f=\rho \times h\times \mathit{dc}/\mathit{dt}\times 273/(273+T)$$

(2)

where $f$ (mg m⁻² h⁻¹) is the N₂O flux, $\rho$ (kg m⁻³) is the density of N₂O under standard conditions, $h$ (m) is the height of the static chamber, $\mathit{dc}/\mathit{dt}$ (mg kg⁻¹ h⁻¹) is the rate of change in N₂O concentration within the chamber per unit time, 273 is the absolute temperature correction factor, and $T$ (°C) is the mean air temperature inside the chamber at the time of sampling.

The cumulative N₂O fluxes during the winter wheat growing season were estimated using the following equation [25]:

$$F={\displaystyle \sum _{i=1}^n\left(\frac{f_i+f_{i+1}}{2}\right)}\times \left(t_{i+1}-t_i\right)\times 24$$

(3)

where $F$ (kg ha⁻¹) is the cumulative N₂O flux, $i$ is the $i$-th sampling time, and $\left(t_{i+1}-t_i\right)$ (d) is the number of days between two consecutive sampling events.

2.3.3. Measurement of NH₃ Volatilization

NH₃ volatilization in the experimental field was measured using the continuous airflow chamber method. The chamber consisted of a bottomless polyvinyl chloride (PVC) cylinder with an outer diameter of 16 cm, an inner diameter of 15 cm, and a height of 10 cm. Two sponges (16 cm in diameter, 2 cm in thickness) soaked with 15 mL of glycerol phosphate solution were placed inside the chamber to capture NH₃ gas (Figure 2). After wheat sowing, the chambers were placed at the center of each plot. Gas sampling was performed once daily during the first week after fertilization, every two days in the second week, every three days in the third week, and approximately every five days thereafter until NH₃ volatilization was no longer detectable. For each sampling, the chamber was inserted 3 cm into the soil. After sampling, the sponge at the bottom of the chamber was removed and immediately sealed in a plastic bag, and a new sponge was placed in the chamber. The collected sponges were promptly returned to the laboratory for extraction, and the NH₄⁺-N concentration in the extracts was determined using a continuous flow analyzer (Autoanalyzer 3, SEAL, Germany). The NH₃ volatilization flux was calculated according to the following equation [26]:

$$N_{\mathrm{v}}={\displaystyle \frac{M}{S\times D\times 100}}$$

(4)

where $N_{\mathrm{v}}$ (kg N ha⁻¹ d⁻¹) is the soil NH₃ volatilization rate, $M$ (mg) is the amount of NH₄⁺-N absorbed in the sponge, $S$ (m²) is the cross-sectional area of the chamber, and $D$ (d) is the number of days the sponge was exposed in the chamber. The cumulative NH₃ volatilization was calculated as the sum of the product of NH₃ volatilization rate and collection days for each sample.

2.3.4. Leaf Area Index, Dry Matter Accumulation, and Plant N Measurements

Leaf area index (LAI) was measured using a canopy analyzer (SunScan, Delta-T, Burwell, UK) at the jointing, booting, anthesis, 15 days after anthesis, and maturity stages of winter wheat. On clear and cloudless days, the probe of the instrument was placed at the bottom of the wheat canopy to record LAI values.

At the wintering, jointing, booting, anthesis, and maturity stages of winter wheat, 20 representative plants were collected from each plot. At the wintering and jointing stages, whole plants were sampled; at anthesis, plants were separated into leaves, stems + sheaths, and spikes; and at maturity, plants were separated into grains, leaves, stems + sheaths, and glumes + rachis. Samples were first oven-dried at 105 °C for 30 min to inactivate enzymes, then dried at 75 °C to a constant weight, while grains were dried separately at 65 °C. The dry matter was then determined.

After drying, plant organs were ground using a plant mill, and their N concentrations were determined by the H₂SO₄–H₂O₂ digestion method followed by an automatic Kjeldahl analyzer (VAPODEST 200, Gerhardt, Königswinter, Germany). The N accumulation of each organ was calculated as the product of its N concentration and dry matter weight.

2.3.5. Grain Yield and Yield Component Measurement

At maturity, five uniform 0.667 m² quadrats were selected from each plot to measure grain yield. In addition, a 20 m² area was harvested in each plot to determine yield components, including spike number per unit area, grains per spike, and thousand-grain weight. Grain yield was adjusted to a standard moisture content of 12.5% according to the national standard of China (GB 1351-2023) [27].

2.3.6. Calculation of Harvest Index, N Harvest Index, and NUE

Harvest index and N harvest index were calculated using the following equations [28]:

$$\mathit{HI}={\displaystyle \frac{\mathit{Yield}}{\mathit{Biomass}}}\times 100\%$$

(5)

$$\mathit{NHI}={\displaystyle \frac{\mathit{GrainN}}{\mathit{BiomassN}}}\times 100\%$$

(6)

where $\mathit{HI}$ is the harvest index (HI), $\mathit{Yield}$ (kg ha⁻¹) is the grain yield of winter wheat, $\mathit{Biomass}$ (kg ha⁻¹) is the dry matter of winter wheat, $\mathit{NNI}$ is the N harvest index (NHI), $\mathit{GrainN}$ (kg ha⁻¹) is the N accumulation in wheat grain, and $\mathit{BiomassN}$ (kg ha⁻¹) is the total N accumulation in wheat plants.

NUE was calculated using the following equation [29]:

$$\mathit{NUE}={\displaystyle \frac{Y_1-Y_2}{N_{\mathit{applied}}}}$$

(7)

where $\mathit{NUE}$ is the nitrogen use efficiency (NUE), $Y_1$ (kg ha⁻¹) is the total dry matter of winter wheat under N fertilization, $Y_2$ (kg ha⁻¹) is the total dry matter of winter wheat without N fertilization, and $N_{\mathit{applied}}$ (kg N ha⁻¹) is the amount of N fertilizer applied.

2.3.7. N Balance Calculation

N inputs consisted of four components: atmospheric N deposition (wet and dry), non-symbiotic N fixation, fertilizer N, and seed N content. The calculation formula was as follows [30]:

$$N_{\mathit{input}}=N_f+N_a+N_{n-s}+N_s$$

(8)

where $N_{\mathit{input}}$ (kg N ha⁻¹) is the total N input in winter wheat fields, $N_f$ (kg N ha⁻¹) is fertilizer N, $N_a$ (kg N ha⁻¹) is atmospheric N deposition, $N_{n-s}$ (kg N ha⁻¹) is non-symbiotic N fixation, and $N_s$ (kg N ha⁻¹) is seed N content.

Atmospheric N deposition: During the winter wheat growing season, precipitation samples were collected after each rainfall event from the on-site meteorological station, and inorganic N concentrations were determined using a continuous flow analyzer (Autoanalyzer 3, SEAL, Germany). The N input from precipitation was calculated as the product of rainfall amount and inorganic N concentration, and the sum of N inputs over the growing season was defined as atmospheric wet N deposition. In eastern China (including Shanxi Province), the annual atmospheric dry N deposition ranged from 20 to 45 kg N ha⁻¹ yr⁻¹ during 2011–2020, with future levels estimated at approximately 35 kg N ha⁻¹ yr⁻¹ [31]. Therefore, in this study, the atmospheric dry N deposition was set to 35 kg N ha⁻¹.

Non-symbiotic N fixation: For non-leguminous crops, non-symbiotic N fixation generally ranges from 4.5 to 20 kg ha⁻¹ yr⁻¹. Based on the latest findings of Guo et al. [32], the non-symbiotic N fixation of winter wheat in this study was set to 15 kg ha⁻¹ yr⁻¹.

Seed N content: Before sowing each year, seed samples were collected, and seed N concentrations were determined using the H₂SO₄–H₂O₂ digestion method with an automatic Kjeldahl analyzer (VAPODEST 200, Gerhardt, Germany). Seed N content was calculated as the product of seed N concentration and seeding rate.

N outputs: N outputs included harvested N removal, ammonia volatilization N loss, nitrous oxide N loss, N gas emissions, surface runoff, soil erosion, and N leaching. Since the experimental site was flat and no surface runoff or soil erosion was observed during the winter wheat growing seasons, these two components were excluded. The N output was calculated according to the following equation [33]:

$$N_{\mathit{output}}=N_r+N_l+N_v+N_d+N_n$$

(9)

where $N_{\mathit{output}}$ (kg N ha⁻¹) is the total N output from winter wheat fields, $N_r$ (kg N ha⁻¹) is harvested N removal, $N_l$ (kg N ha⁻¹) is N leaching, $N_v$ (kg N ha⁻¹) is ammonia volatilization N loss, $N_d$ (kg N ha⁻¹) is nitrous oxide N loss, and $N_n$ (kg N ha⁻¹) is N gas emissions.

N leaching: At a depth of 200 cm in each plot, a drainage pan (19 cm in diameter, 5 cm in height) was installed to collect soil leachates (Figure 2). Soil leachates were collected on the 1st, 2nd, 3rd, and 5th days after each irrigation and rainfall event using a vacuum pump. The leachates were transported to the laboratory, where their N concentrations were determined using a continuous flow analyzer (Autoanalyzer 3, SEAL, Germany). N leaching was calculated using the following equation [34]:

$$N_l={\displaystyle \sum _{\mathrm{i}=1}^n\frac{C_{\mathrm{i}}\times V_{\mathrm{i}}}{S}}\times 0.01$$

(10)

where $N_l$ (kg N ha⁻¹) is the amount of N leaching, $C_{\mathrm{i}}$ (mg L⁻¹) is the nitrate concentration in the leachate, $V_{\mathrm{i}}$ (L) is the volume of leachate, $\mathrm{n}$ is the number of leachate samples, and $S$ (m²) is the cross-sectional area of the drainage pan.

N gas emissions: Globally, the N₂/N₂O ratio ranges from 2.2 to 4.6, and it is projected to be approximately 3.5 by 2050 [35]. Therefore, in this study, N₂ emissions were estimated as 3.5 times the N₂O emissions.

The calculation formulas for net N surplus and N balance were as follows [36]:

$$N_{\mathit{N S}}=N_{\mathit{input}}-N_r$$

(11)

$$N_B=N_{\mathit{input}}-N_{\mathit{output}}-\mathrm{\Delta }TN$$

(12)

where $N_{\mathit{N S}}$ (kg N ha⁻¹) is the net N surplus, $N_r$ (kg N ha⁻¹) is harvested N removal, $N_B$ (kg N ha⁻¹) is the N balance, and $\mathrm{\Delta }TN$ (kg N ha⁻¹) is the change in total soil N storage between pre-sowing and post-harvest.

2.4. Statistical Analysis

A one-way analysis of variance (ANOVA) was performed to determine the significant differences (p < 0.05) among treatments in cumulative N₂O emissions, cumulative NH₃ emissions, yield, and its components. Pearson correlation coefficients were calculated to analyze the relationships between soil NO₃⁻-N and NH₄⁺-N concentrations, dry matter, LAI, NH₃, N₂O, NUE, and yield at different growth stages of winter wheat. Statistical analyses, including ANOVA and least significant difference (LSD) tests, were conducted using SPSS 27 (SPSS Inc., Chicago, IL, USA). Figures were generated using Origin 2025b (Origin Lab Inc., Northampton, MA, USA), and tables were prepared using Excel 2021 (Microsoft, Redmond, WA, USA).

3. Results

3.1. Soil NO₃⁻-N and NH₄⁺-N

Compared with CK, all fertilization treatments significantly increased soil NO₃⁻-N (by 80.26–109.83%) and NH₄⁺-N (by 156.80–240.80%) contents in the 0–100 cm soil profile during the winter wheat growing season (Figure 3; p < 0.05). At the jointing, booting, and anthesis stages, UAN significantly increased NO₃⁻-N content in the 80–100 cm soil layer by 9.45–39.19% compared with NHN and CON (p < 0.05). In contrast, NHN increased NH₄⁺-N content in the 0–100 cm soil profile by 10.59–28.24% at the jointing stage compared with the other N forms, although the differences were not significant (p > 0.05).

Compared with NON, NHN, and CON, UAN significantly increased soil NO₃⁻-N residues in the 0–100 cm soil profile at maturity by 4.51–11.38% (Figure 4; p < 0.05). UAN also resulted in the highest soil NH₄⁺-N residues at maturity among all N fertilizer forms, although the differences from NHN and CON were not significant (p < 0.05).

3.2. N₂O Flux and NH₃ Volatilization

For the NON and UAN treatments, N₂O fluxes peaked 2–3 days after fertilization, whereas the peak occurred 4–5 days after fertilization under the NHN treatment and 5–6 days under the CON treatment. During periods without fertilization events in the winter wheat growing season, N₂O fluxes in all fertilization treatments remained at relatively low levels (Figure 5). The peak N₂O fluxes followed the order NON > NHN > CON > UAN. Compared with other N fertilizer forms (NON, NHN, and CON), UAN reduced cumulative N₂O emissions by 10.40–15.64% (p < 0.05).

During the winter wheat growing season, the peak of soil NH₃ volatilization under different N fertilizer forms occurred 4–6 days after fertilization and declined to very low levels after 16 days (Figure 6). Compared with CON, NON, NHN, and UAN significantly reduced cumulative NH₃ volatilization by 4.50–10.49% (p < 0.05). Overall, UAN was more effective than the other N forms in reducing gaseous N (N₂O and NH₃) losses (Figure 5 and Figure 6).

3.3. LAI and Dry Matter

As the winter wheat growth period progressed, the LAI first increased and then decreased, while the rate of change in dry matter increased after the jointing stage (Figure 7). Compared with the CK, fertilization significantly increased LAI and dry matter after the jointing stage (p < 0.05). Relative to other N fertilizer forms (NON, NHN, and CON), the application of UAN significantly increased LAI by 1.50–5.92% at the booting, 15 days after anthesis, and maturity stages, and significantly increased dry matter by 3.57–9.35% at maturity (p < 0.05).

3.4. Winter Wheat Yield and NUE

Fertilization significantly increased winter wheat yield by 36.95–59.54% compared with the CK (

Table 1. Effects of different N fertilizer forms on winter wheat yield and its components, HI, and NHI from 2022 to 2024.

Year	N Fertilizer Forms	Ear Number (104 ha−1)	Grain Number per Ear	1000-Grain Weight (g)	Yield (kg ha−1)	HI (%)	NHI (%)
2022–2023	CK	408.29 e	34.23 c	38.77 b	7624.28 d	51.38 d	49.5 d
	NON	643.77 b	47.24 ab	47.86 a	10,846.34 b	53.94 ab	62.37 b
	NHN	578.31 c	45.98 b	48.34 a	10,459.32 c	52.22 bc	61.58 b
	CON	541.72 d	44.36 b	47.25 a	10,441.34 c	54.76 cd	58.39 c
	UAN	684.37 a	49.57 a	49.69 a	11,247.65 a	53.07 a	65.91 a
2023–2024	CK	384.21 c	33.91 c	37.26 b	7013.54 d	50.51 c	48.77 d
	NON	663.13a	46.28 b	46.53 a	10,475.36 b	52.66 ab	61.29 b
	NHN	561.32 b	46.27 b	46.73 a	10,157.32 c	52.03 ab	60.13 b
	CON	533.73 b	44.51 b	46.33 a	10,146.39 c	53.42 bc	56.87 c
	UAN	674.32 a	49.94 a	47.71 a	11,189.31 a	53.67 a	64.1 a

Lowercase letters within the same column indicate significant differences among different N fertilizer treatments in the same winter wheat growing season at the p = 0.05 (n = 4) level.

, p < 0.05). Application of UAN significantly increased spike number (by 1.69–26.34%) and kernels per spike (by 4.93–12.20%) compared with other N fertilizer forms (NON, NHN, and CON), thereby significantly increasing yield by 3.70–10.28% (p < 0.05). In addition, UAN application significantly increased the NHI and NUE by 4.58–12.88% and 12.14–39.25%, respectively, compared with the other N fertilizer forms (NON, NHN, and CON) (p < 0.05; Table 1; Figure 7).

3.5. N Balance

Application of different N fertilizer forms had a significant effect on the N balance of winter wheat fields (p < 0.05), showing a trend of UAN < NON < NHN < CON (

Table 2. Effects of different N fertilizer forms on N balance of winter wheat from 2022 to 2024.

Year	N Flow	CK	NON	NHN	CON	UAN
2022–2023	N input (kg·N·ha−1)
	Fertilizer N	0	180	180	180	180
	Atmospheric N deposition	45.46	45.41	45.57	45.20	45.59
	Seed N content	2.25	2.25	2.25	2.25	2.25
	Non-symbiotic N fixation	15	15	15	15	15
	Total N input	62.71	242.66	242.82	242.45	242.84
	N output (kg·N·ha−1)
	Harvested N removal	89.37	194.51	179.35	175.28	206.37
	Ammonia volatilization N loss	5.09	14.70	15.24	16.23	14.72
	Nitrous oxide N loss	1.54	3.72	3.56	3.69	3.14
	N gas emissions	5.38	13.04	12.44	12.92	11.00
	N leaching	0.69	3.90	3.33	3.61	3.20
	Total N output	102.06 d	229.88 b	213.92 c	211.74 c	238.42 a
	Soil total N variations (kg·N·ha−1)	−12.69 b	−11.63 ab	−12.38 ab	−14.29 c	−11.25 a
	Net N surplus (kg·N·ha−1)	−26.66 e	48.15 c	63.47 b	67.17 a	36.47 d
	N balance (kg·N·ha−1)	−32.66 e	24.41 c	41.28 b	45.01 a	15.67 d
2023–2024	N input (kg·N·ha−1)
	Fertilizer N	0	180	180	180	180
	Atmospheric N deposition	45.38	45.62	45.40	45.24	45.35
	Seed N content	2.18	2.18	2.18	2.18	2.18
	Non-symbiotic N fixation	15	15	15	15	15
	Total N input	62.56	242.80	242.58	242.42	242.53
	N output (kg·N·ha−1)
	Harvested N removal	76.58	184.56	177.36	172.92	198.37
	Ammonia volatilization N loss	5.59	17.86	18.64	19.48	18.26
	Nitrous oxide N loss	1.43	3.92	3.75	3.88	3.36
	N gas emissions	5.01	13.72	13.13	13.58	11.77
	N leaching	0.59	3.77	3.18	3.57	3.27
	Total N output	89.20 d	223.84 b	216.06 c	213.42 c	235.03 a
	Soil total N variations (kg·N·ha−1)	−10.37 b	−9.24 a	−10.53 b	−13.81 c	−8.61 a
	Net N surplus (kg·N·ha−1)	−14.02 e	58.24 c	65.22 b	69.50 a	44.16 d
	N balance (kg·N·ha−1)	−22.27 e	28.20 c	37.04 b	42.81 a	16.11 d

Lowercase letters within the same column indicate significant differences among different N fertilizer treatments in the same winter wheat growing season at the p = 0.05 (n = 4) level.

). Compared with other N fertilizer forms (NON, NHN, and CON), UAN application significantly reduced the net N surplus and N balance of winter wheat by 24.18–45.70% and 25.64–55.82%, respectively (p < 0.05).

3.6. Correlation Analysis

Cumulative N₂O emissions during the winter wheat growing season were significantly negatively correlated with soil NO₃⁻-N at maturity, dry matter, LAI, NUE, and yield while showing a significant positive correlation with N balance (Figure 8, p < 0.05). Yield was significantly positively correlated with soil NO₃⁻-N at the jointing and booting stages (p < 0.05) and highly positively correlated with soil NO₃⁻-N at maturity (p < 0.01). Yield was significantly negatively correlated with soil NH₄⁺-N at the booting stage and N balance (p < 0.05), and highly negatively correlated with soil NH₄⁺-N at anthesis (p < 0.01).

4. Discussion

4.1. N₂O Fluxes and NH₃ Volatilization Under Different N Fertilizer Forms in Winter Wheat Fields

N₂O emissions from agricultural soils contribute to stratospheric ozone depletion and global warming [37]. In this study, we found that N₂O fluxes in winter wheat fields increased sharply following N fertilizer application and were further increased when irrigation was applied simultaneously with topdressing (Figure 5). Similarly, Lang et al. [13] reported that N₂O fluxes peaked during the initial stages of field wetting by rainfall or irrigation, leading to substantial cumulative emissions. This phenomenon can be attributed to increased soil moisture filling soil pores, which creates anaerobic conditions conducive to denitrifying bacteria, thereby promoting the reduction of soil nitrate to N₂O [38]. Moreover, our results showed that the peak N₂O flux occurred earliest under NON and UAN treatments, followed by NHN and then CON (Figure 5). This pattern is likely related to the transformation processes of urea-based fertilizers: CO(NH₂)₂ first hydrolyzes under soil urease activity to form ammonium carbonate and bicarbonate, which are further decomposed into NH₄⁺. The NH₄⁺ ions are subsequently oxidized to nitrite by Nitrosomonas and then to nitrate by Nitrobacter [39]. Therefore, multiple factors such as soil pH and microbial activity jointly influence N₂O fluxes following fertilization [40]. This study found that the cumulative N₂O emissions during the winter wheat growing season were significantly reduced under UAN application compared with NON, NHN, and CON treatments, while the soil NO₃⁻-N residue at maturity was significantly increased (Figure 4 and Figure 5; p < 0.05). Two possible mechanisms may explain this reduction. First, UAN may inhibit the activities of nitrate reductase and nitrite reductase while enhancing hydroxylamine reductase activity [41]. Second, the NH₄⁺ in UAN is partly adsorbed by soil colloids and partly absorbed by wheat roots, while the NO₃⁻ is largely assimilated and stored in leaf cells [42]. These processes ultimately reduce N₂O generation. In addition, UAN application significantly increased soil NO₃⁻-N residues at maturity compared with other N fertilizer forms (Figure 4; p < 0.05). This can be attributed to the adsorption of NH₄⁺ from UAN by soil colloids, followed by gradual nitrification to NO₃⁻, which continuously supplies N for wheat growth [43]. Overall, UAN application effectively mitigated N₂O emissions while enhancing soil NO₃⁻-N availability, supporting sustainable winter wheat production.

After fertilization, the concentration of soil NH₄⁺ increased sharply, leading to high NH₃ volatilization, which can account for more than 30% of applied N loss [44]. In this study, NH₃ volatilization peaked 4–6 days after fertilization under different N fertilizer forms, with higher peaks following topdressing at the jointing stage than those observed after basal fertilization (Figure 6). However, the overall NH₃ volatilization measured in this trial was relatively low (Table 2), likely because supplementary irrigation during topdressing raised soil water content to 80%. Previous studies have shown that N loss decreases with increasing irrigation, and maintaining soil water at 70–80% effectively enhances water use efficiency and reduces gaseous losses from winter wheat fields [22]. Moreover, we observed that UAN application significantly reduced cumulative NH₃ volatilization by 2.04–9.33% compared with NHN and CON, although differences from NON were not significant (Figure 6; p < 0.05). This reduction can be explained by the rapid urease hydrolysis of urea in CON (CO(NH₂)₂ + 3H₂O → CO₂ + 2NH₃ + 2H₂O), which produces large amounts of NH₃ [45], and the reversible reaction between NH₄⁺ and OH⁻ in NHN that increases NH₃ volatilization when soil NH₄⁺ is abundant. Appropriate irrigation mitigates this effect by suppressing NH₃ emissions [46].

4.2. Yield and NUE Under Different N Fertilizer Forms in Winter Wheat

Rational application of N fertilizer is a crucial strategy to ensure high wheat yields, reduce costs, and minimize environmental pollution. In this study, application of UAN significantly increased the number of spikes and grains per spike during the winter wheat growing season compared with other N fertilizer forms, thereby enhancing grain yield (Table 1; p < 0.05). The superior performance of UAN can be attributed to its ability to meet the preference of winter wheat for NO₃⁻-N while simultaneously reducing the risk of excessive NO₃⁻-N leaching. Nitrates and nitrites in the soil suppress nitrification, and the persistent nitrification inhibition effect of UAN significantly increased soil NH₄⁺-N accumulation, thereby improving NUE and yield [17], which is consistent with the findings of Brodowska et al. [46]. Previous studies have also demonstrated that N fertilizer form significantly influences spike number, and that spikelet number and thousand-kernel weight are the key yield determinants in ammonium nitrate fertilization [47]. However, Hussain et al. [48] reported that amide N fertilizer significantly increased wheat yield compared with nitrate, ammonium, and ammonium nitrate fertilizers, which contrasts with our results. A possible explanation lies in the experimental conditions: in our study, topdressing was applied via drip fertigation at the jointing stage. Before hydrolysis, only about 20% of amide N can be adsorbed by soil colloids, and irrigation during topdressing reduced soil temperature, thereby decreasing urease activity [49]. Consequently, amide N fertilizer could not provide sufficient N during the critical demand stages of winter wheat, resulting in reduced yield.

In this study, application of UAN increased the LAI and biomass accumulation at maturity compared with NON, NHN, and CON (Figure 7; p < 0.05). Maintaining higher post-anthesis biomass accumulation is recognized as a key pathway for increasing wheat yield, and both N application rate and fertilizer form exert significant regulatory effects on biomass accumulation and translocation [50]. Our results further demonstrated that soil NO₃⁻-N content at the jointing and booting stages was significantly and positively correlated with yield (p < 0.05), while NO₃⁻-N content at maturity was extremely significantly and positively correlated with yield (Figure 8; p < 0.01). The mechanism behind this lies in the properties of UAN: NH₄⁺-N is readily adsorbed onto soil colloids in the surface layer and subsequently nitrified, thereby ensuring N availability for later growth stages; meanwhile, NO₃⁻-N moves moderately downward after application, enabling the deeper roots of winter wheat to access N [46]. In contrast, soil NH₄⁺-N content at the booting stage showed a significant negative correlation with yield (p < 0.05), and at the anthesis stage showed an extremely significant negative correlation with yield (p < 0.01). This can be explained by the fact that NH₄⁺-N tends to bind strongly to soil particles and remains concentrated in the surface plow layer, where it cannot be effectively absorbed by the deeper root system. Furthermore, excessive NH₄⁺-N accumulation around the roots after the jointing stage can lead to ammonium toxicity, inhibiting root respiration and the uptake of other ions, thereby restricting aboveground growth [51].

In this study, application of UAN significantly increased the NHI and NUE compared with NON, NHN, and CON (Figure 7; Table 1; p < 0.05). Moreover, NUE showed an extremely significant positive correlation with soil NO₃⁻-N concentrations at the jointing, booting, anthesis, and maturity stages (Figure 8; p < 0.01). After the jointing stage, which is a critical period for yield formation, a sufficient but balanced supply of N in the soil is essential. UAN contributes to this by enabling moderate downward movement of NO₃⁻-N under irrigation, which increased NO₃⁻-N availability in the 80–100 cm soil layer (Figure 3). Simultaneously, irrigation enhanced soil moisture, promoting nitrification and accelerating the conversion of NH₄⁺-N in the surface soil into NO₃⁻-N, at which point wheat reaches its peak efficiency in NO₃⁻-N uptake [52].

4.3. N Balance Under Different N Fertilizer Forms in Winter Wheat

N balance refers to the relationship (difference) between N inputs and N outputs within a given system (agroecosystem) over a defined period, and it serves as an important indicator for N management and policy making [53]. N balance reflects either soil N surplus (N input > N output) or N deficit (N input < N output) per unit of cultivated land. In this study, fertilizer application was identified as the main source of N input, accounting for 77.50–77.53% of total N inputs, while harvested N removal contributed 82.40–87.73% of total N outputs (Table 2). Generally, fertilizer N is lost in the forms of NH₃ volatilization, N₂O emissions, runoff, or leaching, which can cause severe environmental problems. Among these, N leaching may lead to groundwater pollution, soil acidification, reduced soil quality, and impaired crop production [54]. In this experiment, the NO₃⁻-N, NH₄⁺-N, and CONH₂-N supplied by UAN exhibited spatiotemporal distribution differences that matched the temporal and spatial patterns of N uptake by winter wheat. This balanced the soil N supply with crop N demand, promoted efficient N utilization, and reduced N losses through N₂O emissions (Figure 5), NH₃ volatilization, and NO₃⁻-N leaching (Figure 6; p < 0.05). Our results showed that application of UAN significantly decreased net N surplus and N balance compared with other fertilizer forms (NON, NHN, and CON) (p < 0.05), respectively. Compared with single-form N fertilizers, UAN markedly enhanced crop N uptake efficiency, reduced net N surplus, effectively lowered gaseous N losses (N₂O and NH₃) that contribute to environmental impacts, and maintained relatively low soil residual NO₃⁻-N concentrations, thereby sustaining field N balance to the greatest extent [55].

Soil moisture serves as the carrier through which fertilizer N enters the N cycle of winter wheat fields. When soil water content is high, NO₃⁻-N in the soil profile tends to migrate downward with water movement. Improper irrigation combined with excessive precipitation can transport inorganic N into deeper soil layers, thereby exacerbating NO₃⁻-N leaching [56]. This process reduces the N supply capacity of the root zone, subjects root growth to nutrient stress, restricts N uptake by crops [57], and ultimately leads to higher net N surplus. On the other hand, although increased soil moisture reduces soil aeration and thus minimizes NH₃ volatilization and associated environmental pollution, the downward movement of NO₃⁻-N increases the soil N surplus, which is unfavorable for maintaining N balance in wheat fields [58]. In the present study, however, supplementary irrigation based on soil moisture monitoring effectively reduced gaseous N losses while ensuring crop N uptake, thereby lowering net N surplus, sustaining N balance in the field, and improving N resource utilization efficiency within the agroecosystem. Furthermore, winter wheat is a deep-rooted crop with a nutrient uptake depth of up to 200 cm, and NO₃⁻-N leached below the root zone (120–200 cm) after the previous wheat harvest can support the growth of the subsequent crop [59]. Application of UAN under irrigation conditions promoted moderate downward movement of NO₃⁻-N, facilitating N uptake and yield formation.

5. Conclusions

The results of this study confirmed our hypothesis that, compared with single-form nitrogen fertilizers (nitrate, ammonium, or amide), the compound-form urea ammonium nitrate (UAN) more effectively improved the leaf area index and biomass accumulation of winter wheat, thereby enhancing yield and nitrogen use efficiency. Meanwhile, UAN application reduced gaseous nitrogen emissions from the farmland ecosystem and significantly decreased both net nitrogen surplus and nitrogen balance. Therefore, integrating high-yield cultivation practices with water–fertilizer management and UAN application provides a feasible strategy for achieving high-yield, high-efficiency, and environmentally sustainable winter wheat production in northern China and similar agroecological regions. However, the large-scale implementation of UAN under water–fertilizer integration systems may be limited by factors such as cost, management complexity, and regional variability in irrigation infrastructure. Future studies should evaluate the economic feasibility and long-term environmental impacts of this practice and explore the combined use of UAN with liquid organic fertilizers or nitrogen stabilizers to further enhance winter wheat productivity and nitrogen balance within agroecosystems.