Optimizing Nitrogen Management in Acidic Tea Orchard Soils: The Role of Biochar-Based Fertilizers in Reducing Losses and Enhancing Sequestration

Abstract

Biochar-based fertilizers have attracted increasing attention as sustainable soil amendments due to their potential to enhance nitrogen (N) retention and mitigate N losses. However, their effects on N dynamics in tea orchard soils remain inadequately understood. This study investigated the impact of biochar-based fertilizer (BF) on N migration and transformation into acidic tea orchard soils through controlled laboratory experiments comprising nine treatments, including sole urea (U) applications and various combinations of BF and U. The results showed that ammonia (NH₃) volatilization peaked within seven days after application. Compared with urea-only treatments, the application of BF at 15 t·ha⁻¹ combined with a low U application rate (0.72 t·ha⁻¹) significantly reduced NH₃ and total dissolved nitrogen losses by up to 22.33% and 33.56%, respectively, while higher BF rates increased these losses. BF applications markedly improved soil N sequestration, as evidenced by increases in total nitrogen, ammonium nitrogen (NH₄⁺-N), nitrate nitrogen (NO₃⁻-N), and the NH₄⁺-N/NO₃⁻-N ratio. Additionally, soil organic carbon, urease activity, and pH were significantly enhanced. Random forest analysis identified soil pH and organic carbon as the primary predictors of NH₃ volatilization and soil N retention. Partial least squares path modeling revealed that the BF-to-urea ratio governed N dynamics by directly influencing N transformation and indirectly modifying soil physicochemical properties. BF applied at ≤15 t·ha⁻¹ with low U inputs exhibited potential for improving N use efficiency and sustainability, pending further field validation.

1. Introduction

Tea (Camellia sinensis) is one of the most widely cultivated crops in tropical and subtropical regions such as China, India, and Sri Lanka [1]. As a key economic crop, it has high nutrient demands, particularly for nitrogen (N), driving tea farmers to apply excessive chemical N fertilizers to maximize yields. In China, the average annual N fertilizer application rate is 533 kg·ha⁻¹, with over 30% of tea plantations experiencing N over-fertilization [2,3]. Unlike annual crops, tea is cultivated as a long-term perennial monoculture, where the same soil is continuously fertilized for decades without crop rotation or fallow periods. This management pattern, coupled with acidic soils (pH 3.5–5.0) and high rainfall conditions typical of tea-growing regions, creates unique challenges for N cycling and retention [4].

The frequent application of N fertilizers, typically administered three to four times per year, combined with the inherently low buffering capacity of tea orchard soils, markedly accelerates nitrification processes and exacerbates nitrate nitrogen (NO₃⁻-N) leaching losses [5]. The pronounced soil acidity characteristic of tea-growing regions further intensifies the solubility and mobility of phytotoxic aluminum ions, while simultaneously suppressing microbial-mediated N transformations [6]. As a result, a substantial proportion of applied N is lost via leaching and gaseous emissions rather than being assimilated by tea plants, resulting in persistently low N use efficiency (often below 40%) [7]. Prolonged and excessive fertilizer inputs have also contributed to soil acidification [8,9], compaction [10], organic matter depletion, and diminished nutrient-use efficiency, thereby amplifying nutrient losses through leaching, surface runoff, and greenhouse gas emissions [11,12]. These chronic nutrient imbalances have progressively impaired soil health, disrupted microbial community diversity and functionality, and constrained the biosynthesis of key biochemical constituents essential for tea yield and quality [3]. Collectively, these issues underscore the pressing necessity to establish sustainable N management practices tailored to the long-term perennial nature of tea plantation systems within the broader framework of global environmental sustainability.

In this context, biochar and fulvic acid have attracted increasing attention as environmentally friendly soil amendments. Biochar is a carbon-rich material produced through the oxygen-limited thermochemical conversion of biomass, which features a highly porous structure [13] that contributes to long-term soil carbon sequestration and reductions in CO₂ emissions [14]. Its unique physicochemical properties, such as enhancing soil cation exchange capacity, buffering soil pH, and improving overall soil structure, can increase soil fertility and crop productivity [15,16,17]. Importantly, biochar also influences the soil N cycle by reducing ammonium nitrogen (NH₄⁺-N) and nitrate nitrogen (NO₃⁻-N) leaching through cation exchange and adsorption [18,19]; mitigates NH₃ and N₂O emissions via microbial regulation [20,21,22]; and improves N fertilizer use efficiency and crop productivity [18,21]. However, its alkalinity may paradoxically promote NH₃ volatilization under certain conditions [23].

Fulvic acid, a short-chain molecule derived from natural humic substances, has high loading capacities and physiological activities [24]. Its application can improve the soil environment, promote plant growth, enhance nutrient uptake efficiency, and increase plant stress resistance, making it a widely used soil amendment [25,26]. Jin et al. [24] found that fulvic acid application elevated soil pH, NH₄⁺-N and NO₃⁻-N contents, and the microbial decomposition ability of soil organic matter. Jiao et al. [27] showed that applying 4.65 g·kg⁻¹ fulvic acid significantly increased tobacco soil organic matter, alkaline nitrogen, pH, and urease activity by 22.04% to 93.16%.

Biochar-based fertilizers (BFs) are organic amendments created by combining biomass-derived biochar with additives such as chemical fertilizers, humic acid, clay, or seaweed. This formulation prolongs nutrient release, improves nutrient-use efficiency, and better meets the long-term nutrient needs of crops [28,29,30]. For instance, Puga et al. [29] demonstrated that a BF formulated with a biochar-to-urea ratio of 2:10 significantly delayed N release and reduced NH₃ volatilization by 14% compared to urea (U) alone. In contrast, Zhou et al. [23] found that while BF increased soil NH₃ emissions relative to conventional fertilizers, they also enhanced N use efficiency and reduced leaching losses. In tea plantations, Yang et al. [31] reported that BF applications improved soil total N, organic matter content, and pH in acidified soils; enhanced microbial diversity; and altered microbial community structure, ultimately improving tea yield and quality. Notwithstanding these advances, the interactive effects of BF and chemical N fertilizer on soil N dynamics, N losses via NH₃ volatilization and leaching, and underlying regulatory mechanisms remain poorly understood in tea orchard ecosystems.

Despite these advances, the mechanisms underlying the interaction between BF and chemical N fertilizers in regulating soil N dynamics in tea orchard soils remain insufficiently understood. Although several studies have explored the effects of biochar or organic amendments on N transformation, most have focused on their independent impacts or on annual crop systems, providing limited insight into the distinctive soil environment of perennial tea plantations [32,33]. The strong acidity, low organic matter content, and continuous fertilization typical of tea soils may alter microbial activity and N conversion pathways, making it necessary to re-evaluate how BF influences N migration and transformation under such conditions.

To address these issues, we formulated a BF by combining biochar with fulvic acid and conducted laboratory incubation experiments. We investigated the effects of varying application rates of this BF in combination with U on N migration and transformation in tea orchard soil, quantified N losses via volatilization and leaching, and identified key factors governing soil N retention. Accordingly, we hypothesize that co-application of BF and U will result in the following: (1) reduce N volatilization; (2) decrease N leaching; and (3) strengthen soil N sinks, thereby collectively minimizing N losses. These findings provide a data basis for the scientific application of BF in sustainable tea plantation management.

2. Materials and Methods

2.1. Soil Collection and Preparation of Biochar-Based Fertilizer

The test soils were collected from a 30-year-old tea orchard in the Shizipu tea plantation of the Anhui Provincial Farming Group, Xuancheng City, Anhui province, China (119.13° E, 30.96° N). Fifteen composite samples were excavated using a spade from the 0–30 cm plough layer. A part of the fresh soil sample was used for basic physicochemical analysis, while the remainder was air-dried, ground to pass through a 2 mm sieve, and stored in a cool, dry place before experimentation. The soil was classified as an acidic clay loam with the following properties: total nitrogen (TN): 1.10 g·kg⁻¹; soil organic carbon (SOC): 12.08 g·kg⁻¹; pH (H₂O): 4.25; NH₄⁺-N: 81.03 mg·kg⁻¹; NO₃⁻-N: 24.36 mg·kg⁻¹; electrical conductivity (EC): 97.61 µS·cm⁻¹; bulk density: 1.15 g·cm⁻³; sand: 25.38%; silt: 40.10%; and clay: 34.52%.

The BF was prepared by carbonizing leguminous straw at 400 °C for 1 h under anaerobic conditions. After carbonization, the biochar was crushed, passed through a 2 mm sieve, and then mixed thoroughly with fulvic acid at a mass ratio of 5:1. The resulting BF has the following characteristics: nitrogen content: 5.3%; carbon content: 55%; C/N ratio: 10.38; and pH: (H₂O, 1:2.5 w/v) 7.3.

2.2. Experimental Design

The experiment comprised nine treatments with three replicates each: a control group (CK), high-urea (U) application (U_high, 0.52 g·kg⁻¹ ≈ 1.50 t·ha⁻¹, dry soil basis), low-U application (U_low, 0.25 g·kg⁻¹ ≈ 0.72 t·ha⁻¹, dry soil basis), and three rates of BF application (BF_0.5, 5 g·kg⁻¹ ≈ 15 t·ha⁻¹; BF₁, 10 g·kg⁻¹ ≈ 30 t·ha⁻¹; BF₃, 30 g·kg⁻¹ ≈ 90 t·ha⁻¹; dry soil basis), each applied in combination with both U_high and U_low. The corresponding application rates were calculated based on the soil layer thickness and a bulk density of 1.15 g·cm⁻³. The specific treatments and application rates are detailed in

Table 1. Urea and biochar-based fertilizer application rate.

Dose (t·ha−1)	CK	Uhigh	BF0.5Uhigh	BF1Uhigh	BF3Uhigh	Ulow	BF0.5Ulow	BF1Ulow	BF3Ulow
U	–	1.50	1.50	1.50	1.50	0.72	0.72	0.72	0.72
BF	–	–	15.00	30.00	90.00	–	15.00	30.00	90.00

Notes: CK, control treatment; U_high and U_low represent 1.50 t·ha⁻¹ and 0.72 t·ha⁻¹ of urea, respectively, while BF_0.5, BF₁, and BF₃ correspond to 15 t·ha⁻¹, 30 t·ha⁻¹, and 90 t·ha⁻¹ of the biochar-based fertilizer, respectively.

. The experimental setup for simulating N migration in tea orchard soil is shown in Figure 1. First, a pot (25 cm in height and 15 cm in inner diameter) was layered with a 2 cm thick layer of quartz sand (particle size 2 mm) at the bottom. Then, it was layered with 5.70 kg of 2 mm sieved soil alone or mixed evenly with BF at different proportions, adjusted to 30% water holding capacity with distilled water, lightly compacted, and incubated indoors at room temperature for one week. Subsequently, 50 g of soil was mixed evenly with different amounts of U and spread on the surface of each treatment soil, followed by coverage with a 0.5 cm thick layer of quartz sand (particle size 2 mm). A 25 cm diameter funnel was placed at the bottom of the pot to collect leachate in a 1 L glass beaker. The entire pot setup was then enclosed within a PVP sealing cover measuring 35 cm (length) × 22 cm (width) × 35 cm (height), and a 200 mL glass bottle containing 0.1 mol·L⁻¹ HCl (pH ≈ 1) was placed adjacent to the pot to capture ammonia (NH₃), resulting in the formation of NH₄Cl [34]. The HCl solution was replaced on days 2, 4, 7, 10, 14, 18, 23, and 28 for NH₃ analysis. Meanwhile, the surface soil pH (0–10 cm) was measured using a portable pH meter (Model ST300, Ohaus Corp., Pine Brook, NJ, USA). The simulation mimicked the monthly average rainfall of 140 mm within 3 months after the application of the topdressing fertilizer in the tea regions of southern China, with a rainfall rate of 18 mm·h⁻¹. Leaching was carried out using 600 mL of distilled water for 2 h every 2 weeks. The beaker was replaced, and the leachate was collected and tested on days 7, 21, 35, 49, 63, 77, and 91. After three months, destructive sampling was conducted. A portion of fresh soil was immediately used to determine NH₄⁺-N, NO₃⁻-N, and urease activity (UE). The remaining soil was air-dried; sieved through a 2 mm mesh; and analyzed for TN, EC, pH, and SOC.

2.3. Analytical Methods

The NH₃ gas captured using the HCl solution, resulting in the formation of NH₄Cl, was analyzed for NH₄⁺-N concentration using a continuous flow analyzer (San++System, Skalar Analytical B.V., Breda, The Netherlands). The NH₃ volatilization rate (F, kg·ha⁻¹·d⁻¹) was calculated according to Equation (1):

$$F={\displaystyle \frac{M}{A\times D}}\times {10}^{-2}$$

(1)

where M is the mass of NH₃ (mg), A is the cross-sectional area (m²), and D is the collection duration (d).

The cumulative NH₃ volatilization loss (E_NH3,tot) (kg·ha⁻¹) was calculated as the sum of all intervals:

$$E_{NH3,tot}={\displaystyle \sum _{i=1}^n{E}_{NH3,i}}$$

(2)

where E_NH3,i is the amount of NH₃ volatilized per unit area during the i-th interval (kg·ha⁻¹), and n is the total number of sampling intervals.

The volume of each leachate was recorded. The EC was measured using a conductivity meter (DDS-307A, Leici, Shanghai, China), and pH was determined using a pH meter (PHS-3C, Leici, Shanghai, China). The concentrations of NH₄⁺-N and NO₃⁻-N in each leachate sample were analyzed using a continuous flow analyzer (San++System, Skalar Analytical B.V., Breda, The Netherlands). Total dissolved nitrogen (TDN) and dissolved organic carbon (DOC) concentrations were measured using a TOC analyzer (TOC-L CPH/CPN, Shimadzu, Kyoto, Japan). Dissolved organic nitrogen (DON) was calculated by subtracting the concentrations of inorganic nitrogen (NH₄⁺-N and NO₃⁻-N) from the TDN.

The total leachate volume (TV) (mL) was calculated according to Equation (3):

$$TV={\displaystyle \sum _{i=1}^n{V}_i}$$

(3)

where V_i is the volume of the i-th leachate (mL), and n is the total number of leaching events.

The leaching TDN, NH₄⁺-N, NO₃⁻-N, DON, and DOC (mg·kg⁻¹, w/w, dry soil basis) were calculated according to Equation (4):

$$L_k={\displaystyle \frac{{\displaystyle {\sum }_{i=1}^n(C_{k,i}\times V_i)}}{m}}$$

(4)

where L_k is the cumulative leaching amount of solute k (mg), C_k,i is the concentration of solute k in the i-th leachate sample (mg·L⁻¹), V_i is the volume of the i-th leachate (L), and n is the total number of leaching events. m is the mass of dry soil (kg). The solutes considered in this study include TDN, NH₄⁺-N, NO₃⁻-N, and DOC.

The TN content of the soil was determined using an elemental analyzer (Vario EL Cube, Elementar, Langenselbold, Germany). The soil NH₄⁺-N and NO₃⁻-N ions were extracted using a 2 mol·L⁻¹ KCl solution at a soil-to-extractant ratio of 1:5 (w/v), and their concentrations were determined using a continuous flow analyzer (San++System, Skalar Analytical B.V., Breda, The Netherlands). Soil organic nitrogen (SON) was calculated by subtracting the concentrations of soil inorganic nitrogen (NH₄⁺-N and NO₃⁻-N) from the TN. SOC was measured using the potassium dichromate oxidation method [35]. Soil urease activity (UE) was measured using the indophenol blue colorimetric method [36]. Soil pH was measured using a pH meter (PHS-3C, Leici, Shanghai, China), with a soil-to-water ratio of 1:2.5. Soil EC was measured using a conductivity meter (DDS-307A, Leici, Shanghai, China), with a soil-to-water ratio of 1:5.

2.4. Statistical Analysis

Data were analyzed using SPSS Statistics 27.0.1. One-way analysis of variance (ANOVA) was performed to assess the effect of pH on NH₃ and the effect of total nitrogen input on soil TN, NH₄⁺-N, NO₃⁻-N, and NH₄⁺-N/NO₃⁻-N. Two-way analysis of variance (ANOVA) was performed to determine the effects of treatments on NH₃, leachate characteristics, and soil parameters. Duncan’s Multiple Range Test (α = 0.05) was applied to assess significant differences among treatment means. At each leaching time, significant differences in TDN, NH₄⁺-N, NO₃⁻-N, and DON concentrations among treatments were analyzed using one-way ANOVA, followed by a post hoc LSD test, performed with the “stats” package in R (version 4.5.0). Random forest analysis was conducted using the “randomForest” package in R (v4.5.0) to identify the key influencing factors of soil TN, E_NH3,tot, and TDN. Additionally, partial least squares path modeling (PLS-PM) was conducted using the “plspm” package in R (v4.5.0) to investigate the direct and indirect effects of BF and U dosage, soil physicochemical properties (soil NH₄⁺-N, soil NO₃⁻-N, NH₄⁺-N/NO₃⁻-N, SOC, pH, UE, EC, and C/N), and leachate characteristics (leachate NH₄⁺-N, leachate NO₃⁻-N, DOC, and TV) on TN, TDN, and NH₃.

3. Results

3.1. Dynamic Changes and Cumulative Amount of Ammonia Volatilization in Tea Orchard Soils

3.1.1. Dynamic Changes and Cumulative Losses of Ammonia Volatilization

The dynamic changes in NH₃ volatilization within 28 days after U application into the soil showed that the NH₃ volatilization rate under the CK treatment was relatively low, without any distinct peak, and a maximum rate of 0.08 kg·ha⁻¹·d⁻¹ (Figure 2a) was observed. In comparison with the CK treatment, all fertilization treatments significantly enhanced NH₃ volatilization, and the volatilization rate showed an increasing and then decreasing trend after fertilizer application. The NH₃ volatilization rates under the U_high treatment group were consistently higher than those under the U_low treatment group. NH₃ volatilization peaked on the 4th day for the U_low, U_high, BF_0.5U_low, and BF_0.5U_high treatments, while it peaked on the 7th day for the BF₁U_low, BF₁U_high, BF₃U_low, and BF₃U_high treatments. The U_low and BF_0.5U_low treatments had the shortest NH₃ volatilization time, which was 18 days. There was no significant change thereafter. However, the BF₁U_high and BF₃U_high treatments experienced NH₃ volatilization that lasted 28 days, and then, there was no significant change.

Compared with the CK treatment, all fertilization treatments significantly increased E_NH3,tot losses (Figure 2b). Under both N levels, E_NH3,tot was lower in the low-nitrogen treatments than in the high-nitrogen treatments. Notably, in both the U_high and U_low treatment groups, the combination of U with a low dosage of BF (BF_0.5) resulted in the lowest E_NH3,tot. Specifically, the E_NH3,tot values of BF_0.5U_low and BF_0.5U_high were 3.72 and 7.51 kg·ha⁻¹, respectively, representing reductions of 22.33% and 7.96% compared to U_low and U_high, respectively. In contrast, the BF₃ treatments significantly increased the E_NH3,tot, with losses reaching 6.73 and 12.44 kg·ha⁻¹, showing increases of 40.29% and 52.33% compared to U_low and U_high, respectively. The combined application of N and BF significantly affected E_NH3,tot in the soil (p < 0.001).

3.1.2. Soil pH Dynamics and Its Relationship with Cumulative Ammonia Volatilization

During the NH₃ collection period, the soil pH for each treatment gradually decreased from 4.75–6.93 to 3.77–5.52 (Figure 2c). The treatment groups with N fertilizer alone did not show significant differences compared to the CK treatment. In contrast, the treatments with the combined application of BF generally had higher soil pH values than both the CK and the nitrogen-only treatments. Moreover, the soil pH increased as the amount of BF increased. Compared with the CK treatment, BF treatments increased the soil’s pH by 0.76–1.75 units.

Linear regression analysis indicated that the NH₃ volatilization rate in soil was significantly positively correlated with soil pH under both high and low N levels (R² = 0.67 and R² = 0.55, respectively; p < 0.001) (Figure 2d). Furthermore, the slope of the regression line was steeper in the U_high treatment group than in the U_low group, indicating that NH₃ volatilization losses were more sensitive to pH changes under high-nitrogen conditions.

3.2. Dynamic Characteristics and Accumulation of Nitrogen Leaching in Tea Orchard Soils

3.2.1. pH and Electrical Conductivity of Leachate

The pH values of the leachate under all treatments exhibited a gradual decrease throughout the leaching periods (Figure 3a). Specifically, the pH in the CK treatment stabilized after the 21st day, while the pH in the U_low and U_high treatments gradually stabilized after the 49th day. The pH of the leachate in the BF treatments was significantly higher than that in the CK treatment and the urea-only treatments. Additionally, as the amount of BF applied increased, the pH correspondingly increased. Compared to CK, the pH of the leachate in the BF treatments increased by 0.31–1.6 units. Similarly, both U and BF treatments resulted in an increase in the EC of the leachate compared to CK, with increments ranging from 1.59 to 8.02 times (Figure 3b). Furthermore, the EC increased as the fertilizer application rate increased, with the BF treatments having a higher overall EC than the urea-only treatments.

3.2.2. Leachate Volume and DOC Content

All treatments, except for the U_low treatment, reduced the leachate volume compared to the CK treatment (3357.88 mL), with reductions ranging from 80.91 to 353.61 mL (Figure 3c). Among these, the BF₃ treatment showed the most remarkable reduction (p < 0.001). Compared to CK, BF₃U_low and BF₃U_high reduced the leachate volume by 10.42% and 10.71%, respectively.

Both BF and U treatments significantly impacted the cumulative content of DOC in the leachate (p < 0.001). The cumulative DOC content increased significantly as the application of BF increased (Figure 3d). Compared to the CK treatment, the cumulative DOC content under the BF treatments increased by 2.48–8.16 times.

3.2.3. Nitrogen Leaching Characteristics

The dynamic changes in the concentration of N in the leachate for different treatments are shown in Figure 4a–d. During the leaching period, significant differences were observed in TDN concentrations among the different treatments (22.37–467.08 mg·L⁻¹) (Figure 4a). TDN concentrations initially increased, peaked between 49 and 77 days after fertilization, and then decreased. The increase in TDN concentrations in the U_high treatment group was generally higher than that in the U_low treatment group. The leaching peak of NH₄⁺-N for CK, U_high, and U_low treatments all occurred on the 7th day after fertilization, with values of 14.75, 36.04, and 25.27 mg·L⁻¹, respectively (Figure 4b). For the BF treatments, the peaks were reached on the 49th day (50.91–111.19 mg·L⁻¹). Subsequently, the NH₄⁺-N concentration gradually decreased in all treatments, but the BF₃ treatment showed a subsequent increase in NH₄⁺-N on the 77th and 91st day. Throughout the leaching period, the U_high treatment group showed a higher increase in NH₄⁺-N than the U_low treatment group. The NO₃⁻-N concentrations in all treatments also followed an initially increasing and then decreasing pattern (Figure 4c). The U_high treatment almost showed higher NO₃⁻-N concentrations in the leachate than the U_low treatment. Among these treatments, the difference between the maximum and minimum values for CK treatment was the smallest (11.10 mg·L⁻¹), while that for the U_high treatment was the largest (147.15 mg·L⁻¹). Both the U_high and U_low treatments reached their peaks of NO₃⁻-N concentrations on day 49, whereas the peaks occurred on day 63 and day 77, respectively, when combined with the BF application. The leaching of DON also differed among treatments (Figure 4d). The CK treatment showed minimal fluctuation, while most of the U_high treatments, except for BF₁U_high, showed no significant peaks. However, the U_low combined with BF treatments exhibited double peaks on day 49 and day 77.

3.2.4. Cumulative Nitrogen Leaching Losses

As shown in

Table 2. Leachate contents of TDN, NH₄⁺-N, NO₃⁻-N, and DON, and the NH₄⁺-N/NO₃⁻-N ratio for biochar-based fertilizer, urea, and their combined treatments.

Treatment	TDN (mg·kg−1)	NH4+-N (mg·kg−1)	NO3−-N (mg·kg−1)	DON (mg·kg−1)	NH4+-N/NO3−-N
CK	23.49 ± 1.22 g	2.13 ± 0.03 g	13.99 ± 0.25 g	7.37 ± 1.02 g	0.15 ± 0.00 f
Ulow	95.09 ± 3.78 e	3.52 ± 0.16 f	47.09 ± 0.93 b	44.48 ± 3.99 ef	0.07 ± 0.00 f
BF0.5Ulow	76.27 ± 1.79 f	19.59 ± 0.85 d	19.27 ± 0.80 f	37.42 ± 2.31 f	1.02 ± 0.06 bc
BF1Ulow	152.32 ± 1.66 c	32.21 ± 1.23 c	30.25 ± 1.60 e	89.87 ± 1.06 c	1.07 ± 0.09 b
BF3Ulow	187.03 ± 11.30 b	43.36 ± 2.57 b	35.27 ± 2.03 d	108.39 ± 9.91 b	1.23 ± 0.03 a
Uhigh	156.80 ± 5.94 c	5.91 ± 0.25 e	84.74 ± 3.87 a	66.14 ± 2.12 d	0.07 ± 0.00 f
BF0.5Uhigh	104.17 ± 3.70 d	19.53 ± 0.70 d	36.67 ± 1.81 d	47.97 ± 3.56 e	0.53 ± 0.02 e
BF1Uhigh	158.19 ± 5.59 c	32.98 ± 0.11 c	42.70 ± 0.79 c	82.51 ± 6.28 c	0.77 ± 0.02 d
BF3Uhigh	213.66 ± 4.55 a	46.36 ± 0.57 a	48.74 ± 0.97 b	118.56 ± 4.62 a	0.95 ± 0.03 c
BF	*	**	ns	*	*
U	*	ns	*	*	ns
BF × U	**	ns	**	**	**

Notes: Values are shown as the mean ± standard errors. Different letters indicate significant differences between treatments (p < 0.05); *, **, and ns represent significant differences, with extremely significant differences and no differences at p < 0.05, p < 0.01, and p > 0.05 levels. TDN, Total dissolved nitrogen; NH₄⁺-N, ammonium nitrogen; NO₃⁻-N, nitrate nitrogen; DON, dissolved organic nitrogen; NH₄⁺-N/NO₃⁻-N, ammonium-to-nitrate ratio.

, both BF and U had significant effects on the cumulative leaching loss of TDN, with a significant interaction between BF and U (p < 0.01). The cumulative leaching loss of TDN in all fertilized treatments was 3.25–9.10 times that of the CK treatment. Among the BF treatments, BF_0.5 exhibited the lowest TDN leaching loss, which decreased by 19.79% and 33.56% compared to the U_low and U_high treatments, respectively. The BF₁ and BF₃ treatments increased TDN leaching losses by 0.89–96.68% compared to the urea-only treatments. The cumulative leaching loss of NH₄⁺-N was significantly affected by BF (p < 0.01). Compared to CK, the application of BF significantly increased the cumulative NH₄⁺-N leaching loss, with the increase being greater as the amount of BF applied increased. Compared to the urea-only treatments, the NH₄⁺-N cumulative leaching loss increased by 3.3–12.31 times. U had a significant effect on NO₃⁻-N leaching, and there was also a significant interaction between U and BF (p < 0.01). Compared to CK, all fertilized treatments promoted an increase in NO₃⁻-N in the leachate. However, when compared to urea-only treatments, the addition of BF significantly reduced the NO₃⁻-N content, with reductions of 25.10–59.09% compared to urea-only treatments. Similarly to TDN, the DON content in the leachate increased in the order of CK < BF_0.5 < U < BF₁ < BF₃. The BF_0.5 treatment showed a reduction in DON by 15.87% and 27.47% compared to U, while the BF₁ and BF₃ treatments increased the DON content by 24.74–143.70% compared to U. The NH₄⁺-N/NO₃⁻-N ratio in the leachate was the lowest in both U_low and U_high treatments at 0.07, while the highest ratio was 1.23 for BF₃U_low. BF significantly affected the NH₄⁺-N/NO₃⁻-N ratio, and there was a significant interaction between BF and U (p < 0.01).

3.3. Soil Nitrogen Forms and Chemical Properties in the Tea Orchard

3.3.1. Effects of Fertilization on Soil Nitrogen Forms

As shown in

Table 3. Soil TN, NH₄⁺-N, NO₃⁻-N, SOC EC, pH, UE NH₄⁺-N/NO₃⁻-N ratio, and C/N ratio under biochar-based fertilizer, urea, and their combined treatments.

Treatment	Soil TN (g·kg−1)	Soil NH4+-N (mg·kg−1)	Soil NO3−-N (mg·kg−1)	Soil NH4+-N/NO3−-N	SOC (g·kg−1)	EC (μs·cm−1)	UE (mg·g−1·24 h−1)	Soil C/N	pH
CK	1.06 ± 0.05 g	64.12 ± 5.00 f	22.35 ± 1.97 e	2.89 ± 0.41 b	11.71 ± 0.64 d	89.58 ± 5.83 g	0.56 ± 0.04 e	11.05 ± 0.22 a	4.17 ± 0.03 d
Ulow	1.08 ± 0.00 fg	58.63 ± 4.88 f	34.24 ± 3.09 d	1.73 ± 0.31 c	11.53 ± 0.11 d	143.21 ± 9.71 fg	0.95 ± 0.07 d	10.63 ± 0.13 ab	4.05 ± 0.02 e
BF0.5Ulow	1.35 ± 0.02 e	180.60 ± 9.24 d	69.23 ± 6.30 c	2.62 ± 0.16 b	13.02 ± 0.34 c	293.70 ± 16.48 e	1.38 ± 0.12 c	9.62 ± 0.13 c	4.46 ± 0.04 bc
BF1Ulow	1.55 ± 0.04 c	262.07 ± 16.29 c	64.65 ± 5.23 c	4.07 ± 0.40 a	14.55 ± 0.48 b	465.40 ± 14.90 d	1.60 ± 0.10 bc	9.36 ± 0.35 c	4.55 ± 0.05 b
BF3Ulow	2.55 ± 0.06 a	357.72 ± 11.26 a	85.64 ± 0.85 b	4.18 ± 0.16 a	20.64 ± 1.06 a	783.25 ± 64.97 b	2.07 ± 0.15 a	8.10 ± 0.50 d	5.00 ± 0.05 a
Uhigh	1.14 ± 0.01 f	102.45 ± 8.45 e	57.59 ± 6.14 c	1.79 ± 0.21 c	11.64 ± 0.58 d	179.30 ± 15.91 f	0.90 ± 0.07 d	10.21 ± 0.49 b	4.01 ± 0.03 e
BF0.5Uhigh	1.46 ± 0.05 d	253.82 ± 11.33 c	92.45 ± 8.25 b	2.76 ± 0.21 b	13.05 ± 0.36 c	352.30 ± 15.83 e	1.44 ± 0.09 c	8.97 ± 0.47 c	4.36 ± 0.07 c
BF1Uhigh	1.63 ± 0.02 b	279.00 ± 4.00 b	158.96 ± 6.20 a	1.76 ± 0.08 c	14.65 ± 0.09 b	568.85 ± 49.54 c	1.69 ± 0.10 b	8.96 ± 0.05 c	4.57 ± 0.04 b
BF3Uhigh	2.61 ± 0.03 a	351.35 ± 3.51 a	150.66 ± 12.77 a	2.34 ± 0.18 b	20.69 ± 0.90 a	947.70 ± 54.34 a	2.12 ± 0.18 a	7.94 ± 0.33 d	5.04 ± 0.1 a
BF	**	**	ns	ns	**	**	**	**	**
U	*	ns	ns	ns	ns	ns	*	*	ns
BF × U	ns	**	**	**	ns	*	ns	ns	ns

Notes: Values are shown as the mean ± standard errors. Different letters indicate significant differences between treatments (p < 0.05); *, **, and ns represent significant differences, with extremely significant differences and no differences at p < 0.05, p < 0.01, and p > 0.05 levels. Soil TN, Soil total nitrogen; soil NH₄⁺-N, soil ammonium nitrogen; soil NO₃⁻-N, soil nitrate nitrogen; soil NH₄⁺-N/NO₃⁻-N, ammonium-to-nitrate ratio; soil C/N, carbon-to-nitrogen ratio; EC, electrical conductivity; UE, urease activity.

, compared to CK, all treatments, except for U_low, significantly increased the soil total nitrogen (TN) content, with increases ranging from 1.89% to 146.23%. The treatments with BF increased soil TN content 1.25 to 2.36 times more than the urea-only treatments. Both BF and U significantly affected soil TN (p < 0.05). Figure 5a indicated that soil TN was positively correlated with N input in both U_low and U_high treatments (p < 0.001). The trend in soil NH₄⁺-N content followed that of TN, with increases ranging from 59.78% to 457.89%, except for U_low. The BF treatment increased soil NH₄⁺-N content 2.48 to 6.1 times more than the urea-only treatments. BF significantly affected soil NH₄⁺-N (p < 0.01), and there was an interaction between BF and U (p < 0.01). Soil NH₄⁺-N content was also positively correlated with N input in both U_low and U_high treatments (p < 0.001) (Figure 5b). For soil NO₃⁻-N content, the BF treatments were 1.61 to 2.76 times higher than the urea-only treatments. For the same amount of BF, the U_high treatment was significantly higher than the U_low treatment. As shown in Figure 5c, soil NO₃⁻-N content was positively correlated with N input (p < 0.001), with the slope for U_high being steeper than for U_low. For the NH₄⁺-N/NO₃⁻-N ratio, the lowest ammonium to nitrate ratios were observed in the U_low and U_high treatments, with values of 1.73 and 1.79, respectively, which were 38.06% to 40.14% lower than in CK. In contrast, the BF₃U_low and BF₁U_low treatments showed significantly higher ratios than the other treatments, with increases of 44.64% to 40.83% compared to CK. Figure 5d showed that the NH₄⁺-N/NO₃⁻-N ratio was significantly positively correlated with N input in the U_low treatment group (R² = 0.50, p = 0.002), but no significant correlation was found in the U_high treatment group (R² = 0.03, p = 0.58).

3.3.2. Impact of Fertilization on Soil Chemical Properties

Compared to CK, the BF treatments significantly increased SOC content, with increases ranging from 11.19% to 76.69%, whereas no significant effects were observed in U_low and U_high treatments (Table 3). BF had a significant effect on SOC (p < 0.01), but no interaction between BF and U was observed. Soil EC was mainly affected by BF (p < 0.01), with an increase in EC corresponding to the amount of BF applied. There was an interaction between BF and U (p < 0.05). Under U_low conditions, EC increased by 0.60 to 7.74 times in BF treatments compared to CK, while under U_high conditions, EC increased by 1.00 to 9.58 times. Soil urease (UE) content was influenced by both BF (p < 0.01) and U (p < 0.05). The U_low and U_high treatments increased UE by 1.61 and 1.70 times, respectively, compared to CK, while treatments with BF increased UE by 2.46 to 3.79 times compared to CK. All fertilization treatments significantly reduced the soil C/N ratio, with urea-only treatments reducing it by 3.8% to 7.6% compared to CK, and BF treatments reducing it by 12.94% to 28.14%. Both BF (p < 0.01) and U (p < 0.05) had significant effects on the C/N ratio, though no interaction was observed between BF and U. At the end of the cultivation period, the soil pH in U_low and U_high treatments decreased by 0.12 and 0.16, respectively, compared to CK, while treatments with BF increased pH by 0.19 to 0.87 compared to CK. BF had a significant effect on soil pH (p < 0.01).

3.4. Analysis of Factors Influencing Nitrogen Migration and Transformation in Tea Orchard Soils

3.4.1. Effects of Fertilization on Soil Nitrogen Accumulation and Loss

The net increase in soil N across the fertilization treatments ranged from 0.08 to 1.55 g·kg⁻¹ (

Table 4. Nitrogen contents in the tea-planted soil of different fertilizer treatments.

N Content	CK	Uhigh				Ulow
		Uhigh	BF0.5	BF1	BF3	Ulow	BF0.5	BF1	BF3
Fertilizer application (g·kg−1)	0	0.24	0.51	0.77	1.83	0.11	0.38	0.64	1.70
Mineral N residue (g·kg−1)	1.06 g	1.14 f	1.46 d	1.63 b	2.61 a	1.08 g	1.35 e	1.55 c	2.55 a
Losses via leaching and NH3 volatilization (mg·kg−1)	24.40 h	165.80 c	112.46 e	169.96 c	227.37 a	100.37 f	80.37 g	158.00 d	194.44 b
NH3 volatilization loss (mg·kg−1)	0.91 g	9.00 c	8.28 c	11.77 b	13.71 a	5.28 e	4.10 f	5.68 e	7.41 d
Leaching loss of total N (mg·kg−1)	23.49 g	156.80 c	104.17 d	158.19 c	213.66 a	95.09 e	76.27 f	152.32 c	187.03 b

Notes: Values are shown as the mean. Different letters indicate significant differences between treatments (p < 0.05).

), representing 18.73–87.67% of the total N input (

Table 5. Fate of nitrogen in the tea-planted soil of different fertilizer treatments.

Ratio * of Applied N (%)	Uhigh				Ulow
	Uhigh	BF0.5	BF1	BF3	Ulow	BF0.5	BF1	BF3
Minerals N residue	33.37 b	77.84 a	74.68 a	84.54 a	18.73 bc	77.46 a	77.29 a	87.67 a
Losses via leaching and NH3 volatilization	58.91 b	17.27 cd	18.90 c	11.09 e	69.06 a	14.73 d	20.88 c	10.00 e
NH3 volatilization loss	3.37 b	1.44 c	1.41 c	0.70 d	3.97 a	0.84 d	0.74 d	0.38 e
Leaching loss of total N	55.55 b	15.82 d	17.49 cd	10.39 e	65.10 a	13.89 d	20.13 c	9.62 e
Undefined #	7.61 b	4.89 d	6.42 c	4.37 d	12.21 a	7.82 b	1.83 f	2.33 e

Notes: Values are shown as the mean. Different letters indicate significant differences between treatments (p < 0.05). * Ratio = (amount in fertilizer applied treatment − amount in control treatment)/N fertilizer applied × 100%. ^# Undefined = 100% − ratio of mineral N residue (%) − ratio of losses via leaching and NH₃ volatilization.

). BF additions significantly enhanced the net N accumulation in the soil. Specifically, in the U_low treatment group, the proportion of net N increase in treatments with BF was 4.14–4.68 times higher than U_low alone, while in the U_high group, it was 2.33–2.53 times greater than in U_high treatment without BF (Table 5). At the end of the soil incubation period, the net N losses (mainly through NH₃ volatilization and N leaching) in U_high and U_low treatments were 141.40 and 75.97 mg·kg⁻¹ (Table 4), respectively, accounting for 58.91% and 69.06% of the total N input (Table 5). Within these treatments, the lowest N loss occurred in the BF_0.5 treatment, with net losses of 88.05 and 55.97 mg·kg⁻¹ for U_high and U_low, respectively (Table 4). Compared to U_high and U_low alone, BF addition reduced the proportion of net nitrogen loss by 67.92–85.52% (Table 5). Among the N loss pathways, leaching was the dominant route, accounting for 9.62–65.10% of net N losses (Table 5). The highest leaching losses were observed in the urea-only treatments (55.55–65.10%), whereas BF incorporation reduced N leaching by 38.06–55.48% relative to urea-only applications (Table 5). E_NH3,tot accounted for a smaller proportion of net N losses (0.38–3.97%), with the highest values recorded in U_high and U_low treatments (3.97% and 3.37%, respectively). The addition of BF effectively reduced E_NH3,tot losses by 1.93–3.59% compared to urea-only treatments (Table 5).

3.4.2. Correlation and Random Forest Analysis of Nitrogen Drivers

Correlation analyses (Figure 6a) revealed that BF was positively correlated with soil TN (p < 0.01). Soil NH₄⁺-N, NO₃⁻-N, NH₄⁺-N/NO₃⁻-N ratio, C/N ratio, SOC, DOC, UE, pH, and EC all showed significant positive correlations with soil TN (p < 0.05), while TV (total leachate volume) was negatively correlated with soil TN. BF and U application rates were significant influencing factors of E_NH3,tot and TDN (p < 0.01). Except for the NO₃⁻-N, NH₄⁺-N/NO₃⁻-N ratio, C/N ratio, and TV, all other factors were significantly positively correlated with E_NH3,tot, while TV showed a significant negative correlation. Similarly, BF and U were also key factors influencing TDN (p < 0.01). All factors, except the NH₄⁺-N/NO₃⁻-N ratio and C/N ratio, were significantly positively correlated with TDN (p < 0.01). Random forest analysis (Figure 6b–d) identified the top predictors: for soil TN, pH (7.79%), SOC (7.37%), and soil NH₄⁺-N (7.25%) were the most important variables (p < 0.05); for E_NH3,tot, the key predictors were soil NH₄⁺-N (15.61%), pH (13.20%), and NH₄⁺-N (12.47%) (p < 0.05); for TDN, DON (19.01%), NH₄⁺-N (10.75%), and NO₃⁻-N (8.28%) were most influential (p < 0.05).

3.4.3. Path Modeling of Nitrogen Transformation Mechanisms

Partial least squares path modeling (PLS-PM) was employed to further elucidate the relationships among soil chemical properties, leachate physicochemical characteristics, soil N retention, leaching, and E_NH3,tot under various levels of BF and U co-application (Figure 7a). The results indicate that BF and U co-application had direct and significant positive effects on soil chemical properties (path coefficient = 0.592), soil TN (0.253), TDN (0.990), and E_NH3,tot (0.926). Soil chemical properties directly and positively influenced the leachate properties (0.398), while showing a significant negative effect on TDN (−0.737). Leachate properties had significant positive effects on both TDN (0.902) and TN (0.682), whereas TN had a direct negative effect on TDN (−0.552). As shown in Figure 7b, the impact of BF and U co-application on leachate properties (0.912) was primarily indirect, and the influence of soil chemical properties on E_NH3,tot (−0.832) was also predominantly indirect. In summary, the proportion and application rate of BF and U were the principal driving factors governing N migration and transformation in tea orchard soils. Their co-application not only directly affected the physicochemical properties of the soil but also indirectly influenced N retention, leaching, and volatilization.

3.4.4. Schematic Diagram of Nitrogen Migration and Transformation in Acidified Tea Orchard Soils

In order to visually summarize the effects of BF and U on soil N dynamics, a schematic diagram was prepared (Figure 8). Among the nine treatments, the representative groups of U_low (0.72 t·ha⁻¹) and U_low combined with three BF application rates (BF_0.5: 15 t·ha⁻¹; BF₁: 30 t·ha⁻¹; BF₃: 90 t·ha⁻¹) were selected for illustration. The U_low treatment served as the reference, while BF_0.5, BF₁, and BF₃ represented low, medium, and high BF addition levels, respectively.

As shown in Figure 8, compared with U_low, BF addition enhanced soil TN, SON, N₄⁺-N, and NO₃⁻-N by 24.83–135.10%, 11.33–112.41%, 2.08–5.10 times, and 1.02–1.50 times, respectively, while pH increased by 0.41–0.95 units and SOC by 12.98–79.05%. Regarding gaseous and leaching losses, BF_0.5 reduced NH₃ volatilization by 22.33%, whereas BF₁ and BF₃ increased it by 7.52–40.29%. Compared with U alone, BF_0.5 decreased TDN and DON leaching by 19.79% and 15.87%, respectively, while BF₁ and BF₃ increased them by 60.18–96.68% and 102.06–143.70%. BF treatments increased NH₄⁺-N and DOC concentrations in leachates (by 4.56–11.32 times and 1.02–5.29 times) but reduced NO₃⁻-N by 25.10–59.09%. Therefore, the co-application of BF_0.5 (15 t·ha⁻¹) with U was more effective in mitigating NH₃ volatilization and N leaching while improving N retention in acidified tea orchard soils.

4. Discussion

4.1. Effect of Biochar-Based Fertilizer on Ammonia Volatilization in Tea Orchard Soils

In all fertilization treatments, NH₃ volatilization gradually peaked between days 4 and 7 (Figure 2a), consistent with rapid U hydrolysis during the initial 3 days that generates abundant NH₄⁺, thereby accelerating NH₃ volatilization [37,38]. However, the peaks in BF₁ and BF₃ occurred later (day 7) than those in BF_0.5 and urea-only treatments, indicating that BF altered the temporal dynamics of NH₃ volatilization. This delay is likely attributable to the slower decomposition of biochar, its capacity to adsorb NH₄⁺, and its influence on soil physical–chemical properties, which collectively mitigate NH₃ release [23,39]. Similar observations have been reported in other BF systems, where enhanced NH₄⁺ retention and moderated pH fluctuations contributed to suppressed volatilization [23]. Importantly, NH₃ volatilization in all treatments approached 0 by day 28, indicating that the first 21 days are the key period for regulating NH₃ volatilization loss from the soil.

This study confirmed that increasing the U application rate significantly enhanced NH₃ volatilization. This effect can be explained by the process of U hydrolysis, in which the rapid ammonization of U elevates soil pH, thereby shifting the equilibrium from NH₄⁺ to gaseous NH₃ [40]. Such results are consistent with previous reports that high N fertilization accelerates N loss pathways by altering soil chemical environments and increasing volatilization risk [41]. Therefore, the baseline effect of U provides a critical reference for evaluating the additional impacts of BF application.

When BF was applied at relatively high rates (30–90 t·ha⁻¹), NH₃ volatilization increased by 7.52–52.33% compared with urea-only treatments (Figure 2b). This result is consistent with the findings of Liu et al. [42], who reported that applying alkaline biochar at rates ≥ 40 t·ha⁻¹ increased NH₃ emissions in acidic soils. In our study, the biochar used (pH 7.3) raised the soil pH by 0.31–0.72 units, which increased OH^- availability and accelerated the conversion of NH₄⁺ into NH₃ [40,43]. Thus, under high BF input, the dominant mechanism driving NH₃ volatilization is the alkalinity-induced elevation of soil pH (Figure 2d). In contrast, the addition of BF at a lower rate (15 t·ha⁻¹) resulted in a reduction in NH₃ volatilization by 7.96–22.33% relative to the U_low and U_high treatments, and this is likely due to enhanced NH₄⁺ adsorption and improved soil cation exchange capacity [33,44]. However, this physicochemical explanation alone may not fully account for the observed dose-dependent pattern, as microbial and ionic processes could also play important roles.

Beyond soil pH regulation, the contrasting effects of low and high BF rates may reflect differences in microbial activity and N transformation pathways [42,45]. Low BF rates likely enhanced microbial immobilization of inorganic N and reduced substrate availability for NH₃ volatilization, whereas high BF rates might have provided more labile organic matter and nutrients, stimulating microbial mineralization and nitrification processes that increased NH₄⁺ and NO₃⁻ accumulation [46]. Additionally, excessive BF input may have altered ion balance (e.g., increased competition among NH₄⁺, K⁺, and Ca²⁺) and disrupted adsorption–desorption equilibria, further facilitating NH₃ release [13,47].

Taken together, these findings reveal a dose-dependent response with a potential threshold-like behavior: At lower BF rates (≤15 t·ha⁻¹), adsorption and cation exchange capacity dominate, resulting in reduced NH₃ volatilization, whereas at higher BF rates (≥30 t·ha⁻¹), soil pH elevation becomes the primary driver of volatilization. Although the absolute amount of NH₃ volatilization loss increased under higher BF application, the relative proportion of N lost declined because the legume-derived BF contained 5.3% N, increasing the total N input (Table 4). Therefore, applying 15 t·ha⁻¹ BF appears to be optimal under the present experimental conditions, as it balances soil N retention and NH₃ emission mitigation. Nevertheless, the generalizability of these findings may be affected by soil type, climatic conditions, and BF properties; thus, further studies across diverse agroecosystems are needed to validate this pattern.

4.2. Effects of Biochar-Based Fertilizer on Nitrogen Leaching in Tea Orchard Soils

Biochar is generally reported to mitigate N leaching by adsorbing NH₄⁺ onto surface functional groups and retaining it through cation exchange [44,46]. In contrast, our results showed that BF application markedly increased NH₄⁺-N leaching, with concentrations 3.3–12.31 times higher than those under urea-only treatments (Table 2). Several factors may account for this discrepancy: (1) The NH₄⁺ adsorption capacity of biochar depends strongly on pyrolysis temperature and feedstock characteristics [47], which, in this case, may have been insufficient for immobilizing the available NH₄⁺. (2) Rapid U hydrolysis combined with the mineralization of organic N in BF likely produced more NH₄⁺ than the biochar could retain, thus increasing the NH₄⁺-N content in the leachate. (3) The application of BF increased salt ion concentrations in the soil. When these concentrations surpassed a threshold, nitrification was inhibited, resulting in the accumulation and subsequent leaching of NH₄⁺-N [48]. (4) The source of NH₄⁺ may have shifted over time, initially dominated by U hydrolysis and later supplemented by progressive mineralization of BF-derived N, resulting in persistently high NH₄⁺ levels in the leachate [49]. Taken together, these mechanisms likely explain the unexpected increase in NH₄⁺ leaching observed in the BF treatments.

In contrast, BF application substantially reduced NO₃⁻-N leaching by 25.10–59.09% compared with U alone (Table 2), consistent with previous findings by Huang et al. [50]. This reduction may result from two complementary processes: (1) The partial retention of NH₄⁺ via biochar (mainly through cation exchange and surface adsorption) reduces substrate availability for nitrification, thereby limiting NO₃⁻-N leaching [44]. (2) The elevated EC following BF addition further suppresses nitrification [51], collectively limiting NO₃⁻ formation and mobility. The simultaneous increase in NH₄⁺ and decrease in NO₃⁻ leaching highlights a shift in N species dynamics, driven by the physicochemical effects of biochar on the soil environment.

The dynamic changes in DON concentrations in the leachate further illustrate the complex influence of BF. In the U_low treatment group, the addition of BF exhibited a bimodal trend, with peaks occurring at 49 and 77 days (Figure 4d). This pattern may be attributed to the initial peak resulting from the leaching of small-molecule monomers such as amino acids and U, while the second peak corresponds to larger molecular aggregates, such as humic substances and peptides. By contrast, in the U_high treatment group, no significant peaks were observed in any treatment, possibly because the large amount of U provided ample N, which maintained a relatively stable organic N concentration in the soil solution, rendering it difficult to form significant peaks [52,53]. In terms of DON content in the leachate, the BF_0.5 treatment reduced DON release by 15.87–27.47% compared with U alone (Table 2), showing a similar trend to the 39–56% reductions reported by Shi et al. [54] under biochar–urea co-applications. In contrast, DON leaching was enhanced under BF₁ and BF₃ treatments, possibly due to the higher organic N content in the soybean-straw-derived BF and the additional labile carbon provided by the biochar, which stimulated microbial activity and promoted the mineralization and leaching of organic N. This underscores that BF dosage and composition are critical determinants of DON dynamics.

In this study, the TDN results indicated that BF played a dual role in regulating nitrogen loss, showing both beneficial and adverse effects depending on the application rate. Low BF addition (BF_0.5U_high and BF_0.5U_low) reduced TDN by 33.56% and 19.79%, respectively, compared with U alone, consistent with previous findings that appropriate BF rates enhance soil N retention through the adsorption of NH₄⁺ and suppression of nitrification [55,56]. At low application rates, BF appeared to enhance soil cation exchange capacity and provided suitable conditions for microbial immobilization of inorganic N, thereby decreasing N leaching.

However, excessive BF input (30–90 t·ha⁻¹) promoted TDN leaching (Table 2), suggesting that higher dosages may disturb soil physicochemical balance. The intrinsic N and labile carbon contained in BF can stimulate the mineralization of both BF-derived and native soil organic N, increasing the availability of mobile inorganic N [57]. Moreover, high BF rates can increase soil EC and alter ion competition, weakening NH₄⁺ adsorption and enhancing its leaching potential [13,48]. Overall, these findings indicate that the effect of BF on N loss is dose-dependent, governed by the combined influence of its nutrient composition, impact on microbial activity, and modification of soil ionic balance. The simulated soil layer in this study closely corresponds to the typical 30 cm fine-root distribution zone in tea orchards [58]. The BF_0.5 treatment significantly reduced N leaching. Therefore, this study provides a valuable reference for determining the optimal BF application rate and depth under field conditions.

Overall, the BF application altered the balance of N species in leachate by promoting NH₄⁺ retention, suppressing nitrification and NO₃⁻ production, and modifying organic N turnover. Importantly, the co-application of 15 t·ha⁻¹ BF with U optimized these effects, reducing total N leaching losses while enhancing soil N retention capacity (Table 4). These results suggest that the regulation of N leaching by BF depends on both its intrinsic characteristics and the applied dosage in the simulated system.

4.3. Effects of Biochar-Based Fertilizer on Soil Nitrogen and Physicochemical Properties in Tea Orchard Soils

SOC is widely recognized as a key indicator of soil fertility and quality, as it contributes to nutrient and water retention, aggregate stability, and microbial biodiversity [59,60]. In the present study, the application of BF increased SOC by 11.19–76.69% and TN by 27.36–146.23%, consistent with the results of García-Jaramillo et al. [61]. These results highlight the strong capacity of BF, as a carbon-rich amendment, to enhance soil C and N pools. Partial least squares path modeling (PLS-PM) (Figure 7a) further demonstrated that BF application directly influenced soil TN levels, suggesting that increases in SOC may be a major driver of enhanced N retention. Supporting this view, random forest analysis (Figure 6b) identified SOC as the most important predictor of soil TN, consistent with the role of biochar in stabilizing organic N and mitigating TN leaching [62].

In addition, a clear dose-dependent effect was observed: the 90 t·ha⁻¹ BF treatment produced the greatest increase in soil TN, followed by the 30 t·ha⁻¹ and 15 t·ha⁻¹ treatments (Table 3). This trend aligns with the findings of Zhang et al. [63], who reported that higher rates of Moutai lees-derived biochar (2–4%) more effectively promoted N retention in loess soils than lower rates (0.5–1%). Collectively, these results suggest that the contribution of BF to N retention is regulated by the combined effects of their carbon enrichment properties and application rate.

BF application also altered soil pH, increasing it by 0.19–0.87 units (Table 3), likely due to the alkaline nature of biochar and its release of base cations [64,65]. This is important because soil pH is a key regulator of N availability. Whitman et al. [66] reported that increasing soil pH toward neutral conditions (around 6.5) enhances nutrient retention and availability. These findings suggest that, in addition to SOC enrichment, pH adjustments provide an additional pathway through which BF promotes TN accumulation.

The application of BF markedly enhanced soil inorganic N, with NH₄⁺-N increasing 1.48–5.10 times and NO₃⁻-N by 0.61–1.76 times compared with U alone (Table 3). Similarly to previous studies [59,63,67], this improvement is attributed to stimulated N mineralization and urease activity. Notably, NH₄⁺-N consistently dominated over NO₃⁻-N across BF treatments, likely due to the porous structure of biochar, which improves water retention and creates anaerobic microsites that suppress nitrification, while its alkaline nature and release of base cations further regulate microbial transformations [64,65]. The incorporation of fulvic acid may additionally enhance enzyme activity and microbial community shifts, synergistically promoting inorganic N accumulation [24]. Overall, these mechanisms indicate that BF improves not only the quantity but also the form of available N, enriching NH₄⁺-N relative to NO₃⁻-N, thereby supporting more efficient N uptake in tea orchard systems.

Another important aspect of N dynamics concerns the relative proportions of NH₄⁺-N and NO₃⁻-N, as their balance plays a decisive role in plant nutrition. A higher ratio of NH₄⁺-N to NO₃⁻-N is beneficial for N uptake and metabolism in plants [68]. In tea orchard soils, NH₄⁺-N is generally more favorable for tea plant growth than NO₃⁻-N [69]. Ruan et al. [70] reported that tea plants exhibit a higher uptake rate of NH₄⁺-N than NO₃⁻-N, as they are well-adapted to NH₄⁺-rich environments and possess a strong capacity for NH₄⁺ assimilation in their root systems. In our study, the NH₄⁺-N/NO₃⁻-N ratio under treatments combining BF with low U application (U_low) increased by 51.45–142.62% compared with the U_low treatment alone (Table 3).

These findings highlight that BF, especially when applied at lower rates in combination with reduced U input, can enhance soil C and N pools and adjust the NH₄⁺-N/NO₃⁻-N ratio in ways favorable to tea plant nutrition. While the BF_0.5 treatment (15 t·ha⁻¹) exhibited promising results in improving N retention and reducing losses under laboratory conditions, these outcomes should be interpreted with caution. The absence of plant uptake and environmental variability in the experimental setup limits direct extrapolation to field-scale applications.

To ensure the agronomic relevance of these findings, field trials are needed to evaluate BF performance under actual tea-growing conditions, including its effects on yield and tea quality. In addition, broader implementation of BF will require consideration of production costs, biomass availability, and integration with current fertilization practices. Assessing the economic feasibility and scalability of BF is essential for its adoption as a sustainable strategy in the tea orchard.

5. Conclusions

The first 21 days after fertilization represent a critical period for reducing N losses by controlling NH₃ volatilization. The co-application of U with a low dose of BF (BF_0.5 = 15 t·ha⁻¹) effectively reduced NH₃ volatilization. However, medium-to-high application rates (BF₁ = 30 t·ha⁻¹ and BF₃ = 90 t·ha⁻¹) instead promoted NH₃ volatilization. Compared with NH₃ volatilization, the risk of N leaching in tea orchard soils was found to be higher, with NO₃⁻-N and DON being the primary forms of N loss. Under simulated rainfall conditions, BF_0.5 treatment significantly reduced TDN leaching, while BF₁ and BF₃ treatments increased TDN losses. BF treatments enhanced N retention in tea orchard soils by increasing soil TN, NH₄⁺-N, NO₃⁻-N, and the NH₄⁺-N/NO₃⁻-N ratio, along with improvements in SOC, UE, and pH. Throughout the experiment, BF directly affected N retention, leaching, and NH₃ volatilization, and it indirectly influenced N migration and transformation by modifying soil physicochemical properties. The application of BF reduced the proportion of N losses (via leaching and volatilization) relative to N inputs, thereby enhancing soil N accumulation. Among the influencing factors, soil pH was identified as the dominant predictor of NH₃ volatilization, while both pH and SOC were the key predictors of soil TN content. Although BF can effectively enhance soil N retention, application rates exceeding 30 t·ha⁻¹ may pose environmental risks and increase economic costs. Therefore, a BF application rate of ≤15 t·ha⁻¹ in combination with a low U dose (U_low = 0.72 t·ha⁻¹) appears to be a promising strategy for enhancing soil N retention and optimizing the NH₄⁺-N/NO₃⁻-N ratio, creating a more favorable environment for tea plant growth, and also mitigating NH₃ volatilization and N leaching.

This study used controlled laboratory simulations to clarify how BF regulates N migration and transformation in tea orchard soils. However, the incubation approach inherently lacks plant uptake processes and relies on artificial rainfall simulation, which may not fully represent the dynamic interactions occurring under natural field conditions. In an actual tea orchard, fluctuating temperatures, soil moisture, rainfall patterns, and management practices can markedly influence N transport and microbial activity, potentially resulting in different outcomes. Therefore, long-term in situ field trials in hilly and mountainous tea orchards are needed to validate these findings; assess spatial and temporal variability in N cycling; and better understand their interactions with soil chemical, physical, and biological properties to support region-specific and adaptive fertilization strategies.