Interaction Regulation Mechanism of Soil Organic Carbon Fraction and Greenhouse Gases by Organic and Inorganic Fertilization

Abstract

Under conditions of constant total nutrient input, the regulatory mechanisms of soil organic carbon components under gradient replacement ratios of organic materials for chemical fertilizers have not yet been systematically elucidated. This study took “Longjiao No. 2” as the research object, setting up CK (no fertilization), T0 (100% chemical fertilizer application), T20 (80% chemical fertilizer + 20% vegetable waste organic fertilizer), T40 (60% chemical fertilizer + 40% vegetable waste organic fertilizer), T60 (40% chemical fertilizer + 60% vegetable waste organic fertilizer), and T80 (20% chemical fertilizer + 80% vegetable waste organic fertilizer) as treatment groups. This study investigated the changes in soil organic carbon and organic carbon component content at different crop growth stages (seedling stage, budding stage, flowering and fruit-setting stage, and fruiting stage) under different organic matter replacement methods of chemical fertilizer treatments. It analyzed the response of greenhouse gas emissions to different fertilization conditions and assessed the changes in soil carbon pool management indices, as well as the interaction mechanisms between soil nutrients, carbon components, and greenhouse gases. The results showed that the combined application of chemical fertilizer and vegetable residue organic fertilizer significantly affected soil carbon pool dynamics and greenhouse gas emissions: the T60 treatment was the most effective, increasing soil organic carbon components at all growth stages. The soil carbon pool management index (CPMI) during the seedling stage was 21.3% higher than that of the T0 treatment, and the stable carbon pool components (MOC and POC) during the fruiting stage were 18.7–22.4% higher. This application mode reduced the global warming potential (GWP) by 25.6% compared to the T0 treatment throughout the entire growth stage. The CO₂ emissions peaked 19.3% lower during the flowering and fruit-setting stage. Applying organic fertilizer and chemical fertilizer in a 6:4 ratio balanced carbon turnover and sequestration while achieving the highest yield, providing a basis for low-carbon fertilization and increased production in semi-arid regions’ protected agriculture.

1. Introduction

Pepper cultivation is a pivotal component of facility agriculture in China, characterized by high intensity and frequent fertilization; however, excessive reliance on chemical fertilizers has triggered a severe imbalance in soil nutrient cycling, leading to the dual challenges of soil organic carbon (SOC) loss and increased greenhouse gas (GHG) emissions [1,2]. As the largest terrestrial carbon pool, even minor fluctuations in SOC can significantly impact atmospheric CO₂ concentrations, making its stability crucial for mitigating climate change [3,4]. Conversely, agricultural soils are also a major source of GHG emissions, with irrational fertilization practices contributing substantially to global warming [5,6].

The recycling of agricultural organic waste, such as vegetable residues (tail grass), into organic fertilizers presents a promising strategy for sustainable soil management. Tail grass, with its high organic matter and nutrient content, serves as an ideal resource for organic amendments [7,8]. Integrating such organic fertilizers with chemical fertilizers is widely regarded as an effective approach to enhance soil health and reduce environmental footprints [9,10]. This practice not only improves soil structure and fertility but also has the potential to modulate the dynamics of SOC fractions and mitigate GHG emissions [11,12].

Despite existing research, critical knowledge gaps remain. First, most studies have focused on the short-term effects of a single substitution ratio, leaving a systematic understanding of how a gradient of substitution ratios influences SOC fractions across the entire crop growth cycle poorly understood [13,14,15]. Second, the dynamic responses of key SOC fractions—particularly the stable mineral-associated organic carbon (MOC) and particulate organic carbon (POC)—across critical phenological stages (e.g., seedling, budding, flowering, and fruiting stages) are not well-quantified [16,17,18,19]. Finally, the interactive mechanisms between soil nutrients, SOC fractions, and GHG emissions under different substitution ratios remain elusive, hindering our ability to predict the overall carbon sequestration and emission reduction potential [20,21].

To address these gaps, we conducted a field experiment with a gradient of substitution ratios where chemical fertilizers were replaced by tail-grass-based organic fertilizer. The overarching goal of this study was to elucidate the systematic regulation mechanisms of organic–inorganic fertilization on SOC turnover and GHG emissions in facility-based pepper cultivation. Specifically, we aimed to achieve the following: (1) investigate the dynamics of SOC fractions across different growth stages; (2) assess the response of GHG emissions and the carbon pool management index (CPMI) to varying fertilization regimes; and (3) unravel the interaction mechanisms among soil nutrients, carbon components, and GHG emissions. We hypothesized that a balanced substitution ratio would optimize the synchrony between carbon sequestration and emission reduction, ultimately achieving a win–win scenario for yield enhancement and environmental sustainability.

This research provides critical insights into the carbon-friendly recycling of agricultural waste and offers a theoretical basis for formulating precision fertilization strategies that enhance carbon sequestration and reduce emissions in semi-arid facility agriculture.

2. Materials and Methods

2.1. Study Sites

The field experiment was conducted at the Highland Summer Vegetable Science and Technology Courtyard, located in Yuzhong County, Gansu Province (104°15′ E, 35°45′ N). This region features a temperate semi-arid continental climate, characterized by an average annual temperature of approximately 6.0 °C and an annual precipitation of about 400 mm. The study site is situated at an elevation of 2500 m above sea level, with an annual evaporation rate of 1343 mm, and a frost-free period of approximately 150 days. The soil type at the site is loess soil, and the background characteristics of the soil are presented in

Table 1. Soil background values.

Indicator	pH	SOC (g·kg−1)	TN (g·kg−1)	TP (g·kg−1)	TK (g·kg−1)	AP (mg·kg−1)	AN (mg·kg−1)	AK (mg·kg−1)
Value	8.12	8.32	0.56	0.82	28.00	117.40	76.42	237.70

Note: This table shows the soil nutrient content before fertilization. SOC, TN, TP, TK, AP, AN, and AK represent soil organic carbon, total nitrogen, total phosphorus, total potassium, available phosphorus, alkaline-hydrolyzable nitrogen, and available potassium.

. This combination of environmental factors makes it an ideal location for investigating agricultural productivity in semi-arid highland regions.

2.2. Experimental Design

The experiment employed a multi-factor randomized block design, with the total nutrient application rate kept constant. Six treatments were set up: no fertilization (CK); 100% chemical fertilizer alone (T0); 80% chemical fertilizer + 20% vegetable residue organic fertilizer (T20); 60% chemical fertilizer + 40% vegetable residue organic fertilizer (T40); 40% chemical fertilizer + 60% vegetable waste organic fertilizer (T60); and 20% chemical fertilizer + 80% vegetable waste organic fertilizer (T80). Each treatment was replicated three times. The detailed experimental design is shown in

Table 2. Experimental scheme.

Treatment	Chemical Fertilizer			Organic Fertilizer (kg·hm−2)
	N (kg·hm−2)	P2O5 (kg·hm−2)	K2O (kg·hm−2)
CK	0	0	0	0
T0	277.33	487.17	142.42	0
T20	221.87	389.73	113.93	1916.67
T40	166.40	292.30	85.45	3691.67
T60	110.93	194.87	56.97	5541.67
T80	55.47	97.43	28.48	7391.67

Note: This table shows the fertilizer application rates for different experimental treatments. N, P₂O₅, and K₂O are nitrogen fertilizer, phosphorus fertilizer, and potassium fertilizer.

. The application rate of chemical fertilizers is determined based on the planting experience of local farmers. The application rate of organic fertilizer is balanced based on the nitrogen fertilizer application rate of the T0 treatment. Since the N, P, and K content in the organic fertilizer used is fixed and cannot be adjusted to make the nutrient content exactly the same as the T0 treatment by fixing the organic fertilizer application rate, the balance is adjusted based on the N, P, and K content in sequence. It was found that balancing based on N can make the total nutrient application rate more consistent, at which point K is also basically balanced, although there is some difference in P.

The vegetable residue organic fertilizer was provided by Lanzhou Dadi Great Wall Fertilizer Co., Ltd. (3.0% N, 2.10% P₂O₅, 1.07% K₂O, 30% organic matter) from Lanzhou, China, with a pH of 8.5. The applied fertilizers were urea (nitrogen fertilizer), calcium superphosphate (phosphorus fertilizer), and potassium sulfate (potassium fertilizer); nitrogen, phosphorus, potassium fertilizers, and organic fertilizer were all applied as basal fertilizer in a single application before sowing. The experimental variety was “Longjiao No. 2,” with a plot area of 12 m² (4 m × 3 m). Each plot was planted with 2 ridges, a ridge width of 0.8 m, a ridge height of 0.2 m, and a ridge spacing of 0.4 m. Each ridge was planted with two rows of chili plants, maintaining a plant spacing of 0.5 m between chili seedlings, with eight plants per ridge (Figure 1). A plastic mulch covering system was used, with the mulch being 1-meter-wide and 0.05-millimeter-thick ordinary polyvinyl chloride agricultural mulch. Daily management of the chili field followed the conventional field management practices of local farmers.

2.3. Sample Collection and Measurement

2.3.1. Soil Sample Collection and Determination

During the four key phenological stages of the chili pepper growth cycle (May to September 2024)—the seedling stage, budding stage, flowering and fruit-setting stage, and fruit development and harvest stage—soil samples were systematically collected. The soil sampling depth was uniformly set at 0–20 cm; each plot was sampled twice using a five-point sampling method to collect soil samples, which were then divided using the quartering method to obtain representative composite samples for each plot; two sampling events were conducted within each growth stage, with a one-week interval between them. The samples were then naturally air-dried, sieved through a 2 mm nylon mesh, and mixed in equal volumes to achieve uniformity. After processing, the samples were packaged into sterile polyethylene bags and stored in a dark environment at 4 °C within 24 h of collection, awaiting physical and chemical analyses.

Following sample collection, the soil was divided into two groups for analysis. Fresh soil samples were retained for determination of microbial biomass carbon (MBC) and dissolved organic carbon (DOC). The remaining samples were air-dried for subsequent analysis of organic carbon (SOC), easily oxidizable organic carbon (EOC), mineral-associated organic carbon (MOC), and particulate organic carbon (POC).

Soil physical and chemical properties were determined according to standard methods [22]. Soil organic carbon (SOC) was determined by the potassium dichromate oxidation method with external heating. Total nitrogen (TN) was measured by the semi-micro Kjeldahl method. Total phosphorus (TP) was analyzed by alkali fusion followed by molybdenum–antimony spectrophotometry. Available phosphorus (AP) was extracted with 0.5 mol/L NaHCO₃ and determined by molybdenum–antimony–scandium colorimetry. Available potassium (AK) was extracted with 1 mol/L ammonium acetate (NH₄OAc) and measured by flame photometry. Soil pH was determined potentiometrically in a 1:2.5 (w/v) soil–water suspension.

Soil microbial biomass carbon (MBC) was determined by chloroform fumigation–extraction followed by potassium dichromate oxidation [23]. Dissolved organic carbon (DOC) was extracted with deionized water (soil–water ratio of 1:5) and measured after centrifugation [24]. Readily oxidizable organic carbon (EOC) was determined by oxidation with 333 mmol/L K₂MnO₄ [24]. Particulate organic carbon (POC) and mineral-associated organic carbon (MOC) were separated by density fractionation using sodium hexametaphosphate (50 g/L) as a dispersant [25].

2.3.2. Measurement of GHGS

Soil gas CH₄ and CO₂ emissions were collected during the 2024 pepper growing season (May to September), with emission fluxes determined using the static chamber–gas chromatography method. Sampling occurred weekly between 8:00 and 11:00 a.m. (adjusted for rain or snow), with three replicates per treatment. Extreme weather conditions were avoided during sampling, and operational consistency was maintained throughout the experimental period to ensure data comparability. The static chamber was constructed from stainless steel, measuring 0.40 m (length) × 0.40 m (width) × 0.40 m (height), with an internal air volume of 64 L. The base (0.43 m × 0.43 m) was buried and sealed after planting. A small fan on the side facilitated gas mixing, while the top featured sampling ports and thermometer insertion holes.

Gas sampling commenced immediately after the airlock closure. Using a 100 mL airtight syringe (flushed 3–5 times with ambient air on site), gas samples were collected at three time points: 0, 10, and 20 min. A set of parallel samples (six samples total) was collected at each time point, while simultaneously recording the chamber temperature. Gas samples were then transferred to 0.5 L aluminum–plastic composite gas bags (inspected for airtightness and flushed with ambient air prior to use) for storage. Analysis was performed within 24 h using an Agilent 7890B gas chromatograph (Agilent Technologies, Santa Clara, CA, USA).

The CH₄ and CO₂ analysis employed a flame ionization detector (FID) coupled with a nickel catalyst (detecting CO₂ after catalytic conversion to CH₄). The column was an Agilent HP-PLOT Q PT, operated in constant-flow mode (flow rate 6.5 mL/min). Inlet temperature was 250 °C, detector temperature 300 °C, with nitrogen as the carrier gas. The unique advantage of this system lies in its integration of TCD and catalytic FID, enabling analysis of permanent gases (H₂, O₂, N₂, CH₄, CO, and CO₂) and hydrocarbons (C₁–C₆) across a broad concentration range (0.1 ppm to 100%) [7].

2.4. Calculation Formulae

2.4.1. Calculation of Gas Samples

The calculation of gas emission flux was conducted using the following formula [26,27]:

$$F={\displaystyle \frac{\mathrm{M}\mathrm{c}}{22.4\times ∆t}}\times H\times 273\times \left({\displaystyle \frac{C_2}{T_2}}-{\displaystyle \frac{C_1}{T_1}}\right)$$

where F denotes the emission fluxes of CO₂ and CH₄, expressed in mg·m⁻²·h⁻¹; H represents the height of the sampling dark box in meters; and Mc is the relative molecular mass of CO₂. The parameters C₁ and C₂ correspond to the volume concentrations of the gas inside the static chamber at the initial (0 min) and final (20 min) stages of measurement, respectively, expressed in μmol·mol⁻¹; similarly, T₁ and T₂ represent the temperatures inside the static dark box at the start and end of the measurement period. Finally, Δt represents the time difference (in hours) between the 20th and 0th minute.

The calculation formula for the cumulative emissions of soil CO₂ and CH₄ is outlined below:

$$M=\sum \left[\left(F_{N+1}+F_N\right)\times 0.5\times \left(t_{N+1}-t_N\right)\times 24\times {10}^{-2}\right]$$

where M represents the cumulative greenhouse gas emissions during the measurement period (kg·hm⁻²); F represents the greenhouse gas emission flux (mg·m⁻²·h⁻¹); N represents the number of samples; and t_N+1 − t_N represents the time interval between the two sampling events (days).

Based on the cumulative emissions of CH₄ and CO₂ within the annual reproductive cycle, we calculated the total global warming potential (TGWP) of the gas (in CO₂ units, unit: kilograms). Over a 100-year time scale, the warming potential of one CH₄ molecule is 25 times that of one CO₂ molecule.

The calculation formula is as follows:

$$TGWP={GWP}_{{CO}_2}+f_{{CH}_4}\times 25$$

2.4.2. Calculation of the Carbon Pool Management Index

The CK treatment was used as the reference soil, and the carbon pool activity and soil organic carbon (SOC) of the CK-treated soil were used as the L₀ and SOC values of the reference soil. The carbon pool activity (L) under different fertilization treatments, carbon pool activity index (LI), carbon pool index (CPI), and carbon pool management index (CPMI) were calculated. A CPI > 1 indicates an increase in the total carbon pool relative to the control, while a CPI < 1 indicates a decrease. CPMI stands for the carbon pool management index; this is a comprehensive index that integrates both the size (CPI) and the lability (LI) of the soil organic carbon pool—a higher CPMI value indicates a more favorable shift in soil organic matter quality towards a larger and more active (labile) pool, which is beneficial for nutrient cycling and soil health. The mean values of the total organic carbon and carbon pool activity of the CK-treated soils were used as the SOC content and L₀ value of the reference soils [28].

Carbon Pool Index (CPI): CPI = (SOC _sample)/(SOC _reference).

Carbon Pool Activity (L): L = (OC _active)/(OC _inactive).

Carbon Pool Activity Index (LI): LI = (L _sample)/(L _reference).

Carbon Pool Management Index (CPMI): CPMI = CPI × LI × 100.

In the formula, represents the activity of the corresponding carbon pool. The content of active organic carbon is calculated based on the EOC content, and the content of inactive organic carbon is calculated as the difference between the SOC and EOC contents.

2.5. Data Analysis

The data were preprocessed and organized using Excel 2021. All statistical analyses were performed using R software (version 4.1.2). One-way analysis of variance (ANOVA), followed by Tukey’s honestly significant difference (HSD) post hoc test, was used to compare differences in soil organic carbon fractions, greenhouse gas emissions, and carbon pool management indices among different fertilization treatments across growth stages. Pearson’s correlation analysis was applied to examine the relationships between soil nutrients, carbon components, and cumulative GHG emissions. Results are presented as means ± standard errors (SE). Figures were generated using R 4.1.2 and Origin 2024 Pro.

3. Results

3.1. Effect of Fertilization Practices on Soil Total Organic Carbon and Its Fractions

Figure 2 illustrates the effects of different fertilization treatments (CK, T0–T80) on soil total organic carbon (SOC) and its component contents (MOC, POC, DOC, EOC, and MBC) across four growth stages of chili peppers: seedling stage, budding stage, flowering and fruit-setting stage, and fruiting stage.

Throughout the entire growth period, SOC content in all fertilization treatments (T0–T80) was significantly higher than in the unfertilized control (CK). Among these, the T60 treatment (40% chemical fertilizer + 60% organic fertilizer) showed the most pronounced enhancement effect: during the seedling stage and budding stage, SOC content in the T60 treatment was significantly higher than in all other treatments; during the flowering and fruit-setting stage, T60 was significantly higher than T0, T20, and T40; by the fruiting stage, it remained significantly higher than T0 and T20. As growth progressed, differences among the T20–T80 treatments gradually diminished, indicating that the SOC-promoting effect of high organic fertilizer substitution stabilized in the later growth stages.

Regarding mineral-bound organic carbon (MOC), all fertilization treatments exhibited higher MOC content than CK throughout the entire growth period. The T60 treatment was particularly prominent, being significantly higher than other treatments at the seedling stage and consistently higher than CK and low-ratio fertilization treatments (e.g., T0 and T20) at the budding stage, flowering and fruit-setting stage, and fruiting stage. By the fruiting stage, differences among the T20–T80 treatments were no longer significant.

Particulate organic carbon (POC) content was higher in all fertilization treatments than in the CK. The T60 treatment significantly increased POC content, being significantly higher than other treatments during the seedling and budding stages, and also higher than the CK, T0, and T20 treatments during the flowering and fruit-setting stage and fruiting stage. Differences between the T20–T80 treatments gradually diminished in the later growth stages.

Soluble organic carbon (DOC) content was higher in all fertilization treatments than in CK, with T60 treatment again showing the optimal effect. Its DOC content at the seedling and budding stages was significantly higher than other treatments, and it was also higher than CK, T0, and T20 at the flowering and fruit-setting stage and fruiting stage. As the growth process progressed, the differences between the T20–T80 treatments gradually became insignificant.

Easily oxidizable organic carbon (EOC) content was higher in all fertilization treatments than in CK, with overall higher levels during the flowering and fruit-setting stage. The T60 treatment exhibited significantly higher EOC content than CK, T0, and T20 during the seedling stage; remained significantly higher than T0 at the budding stage; and was significantly higher than all other treatments at the flowering and fruit-setting stage. By the fruiting stage, although the EOC content in the T60 treatment was higher than CK, T0, and T20, it was significantly lower than that in the T80 treatment. As the growth period progressed, the differences in EOC between the T20–T80 treatments gradually decreased.

Microbial biomass carbon (MBC) was also higher in all fertilization treatments than in CK. MBC content in the T60 and T80 treatments at the seedling stage was significantly higher than in CK, T0, and T20. Starting from the budding stage, MBC content increased with rising organic fertilizer proportion, with the T80 treatment performing best and significantly exceeding other treatments. Upon entering the fruiting stage, MBC differences among treatments (T20–T80) diminished, indicating that the soil microbial community’s response to fertilization tended toward equilibrium.

3.2. Effect of Fertilization Practices on the Proportion of Soil Organic Carbon Fractions

As shown in Figure 3, this study analyzed the effects of different fertilization treatments (CK, T0–T80) on the proportional relationships between soil carbon components and soil organic carbon (SOC) at various growth stages of chili peppers, focusing on five indicators: MOC/SOC, EOC/SOC, DOC/SOC, MBC/SOC, and POC/SOC.

Figure 3a indicates that during the seedling stage, all treatment groups (T0–T80) showed minimal differences from CK (49.00), with values fluctuating between 48.28 and 49.16. At the budding stage, the T0 treatment was slightly lower than CK (47.68), while the T40 treatment recorded the lowest value (44.85). From the flowering and fruit-setting stage to the fruiting stage, MOC/SOC values in fertilized treatments were significantly lower than CK. Specifically, T40 decreased to the lowest value of 33.34 during the fruiting stage, while T60 (36.71) and T80 (35.33) showed slight recoveries compared to T40, indicating weaker suppression of MOC/SOC by T60 and T80 than by T40.

As shown in Figure 3b, CK reached its peak EOC/SOC value (12.84) during the flowering and fruit-setting stage, significantly higher than all fertilization treatments (9.81–12.21). The seedling-stage fertilization treatments (16.60–18.74) were slightly higher than CK (11.61); however, treatments applied during the fruiting stage generally showed lower values than CK (15.10), with T40 exhibiting the lowest value (10.30). During the budding stage, T60 (15.29) and T80 (15.52) were slightly higher than CK (12.84), indicating that organic–inorganic combined fertilization significantly enhances EOC/SOC during this stage.

Figure 3c shows that DOC/SOC values were generally low (0.32–0.47) across all growth stages. The CK treatment exhibited the highest value during the seedling stage (0.45), while fertilized treatments were comparable to CK (0.43–0.47) during this period. By the fruiting stage, T20 and T40 reached the lowest values throughout the growth cycle (0.32), while T60 remained at 0.36—still below CK (0.43)—indicating that T60 exerted less suppression on DOC/SOC than T20 and T40.

Figure 3d indicates MBC/SOC peaked at the fruiting stage in CK (1.62), although T60 (1.73) and T80 (1.97) remained higher than CK. At the budding stage, T60 (2.10) and T80 (2.33) were significantly higher than CK (1.41), reaching the highest values throughout the growth period, indicating that fertilization significantly promoted microbial biomass accumulation during budding. During the flowering and fruit-setting stage, all fertilization treatments (1.17–1.71) were close to CK (1.44), with only T80 slightly higher, indicating that fertilization at this stage had a weaker effect on enhancing MBC/SOC compared to the budding stage.

Figure 3e shows that the POC/SOC value of CK remained stable throughout the entire growth period (51.00–52.67), while fertilization treatments exhibited a significant increase from the flowering and fruit-setting stage to the fruiting stage. T40 reached the highest value of 66.66 during the fruiting stage. During the budding stage, all fertilization treatments (54.53–55.15) were also significantly higher than CK (52.32), indicating that fertilization continuously promotes POC/SOC accumulation. T80 reached 63.41 during the flowering and fruit-setting stage, the highest value for this period, suggesting that increasing the organic fertilizer substitution ratio during flowering particularly enhances POC/SOC.

3.3. Effect of Fertilization Practices on Soil Carbon Pool Management Indices

As shown in

Table 3. Soil carbon pool management index under different fertilization treatments.

Period	Treatment	L	LI	CPI	CPMI
Seedling stage	CK	0.143 ± 0.003 b	1.000 ± 0.032 a	1.000 ± 0.000 c	100.033 ± 1.998 d
	T0	0.151 ± 0.002 b	0.922 ± 0.023 c	1.090 ± 0.026 b	114.230 ± 3.885 c
	T20	0.161 ± 0.002 a	0.979 ± 0.014 ab	1.083 ± 0.012 b	121.753 ± 2.124 b
	T40	0.162 ± 0.001 a	0.95 ± 0.031 bc	1.150 ± 0.010 a	129.777 ± 1.420 a
	T60	0.165 ± 0.01 a	0.988 ± 0.015 ab	1.160 ± 0.026 a	133.173 ± 4.997 a
	T80	0.163 ± 0.008 a	0.949 ± 0.02 bc	1.137 ± 0.015 a	128.577 ± 5.770 a
Budding stage	CK	0.194 ± 0.006 a	1.000 ± 0.032 a	1.000 ± 0.010 d	99.980 ± 2.370 d
	T0	0.179 ± 0.005 c	0.922 ± 0.023 c	1.093 ± 0.006 c	100.903 ± 2.153 c
	T20	0.190 ± 0.007 a	0.979 ± 0.014 ab	1.147 ± 0.021 b	112.127 ± 1.642 c
	T40	0.184 ± 0.006 bc	0.950 ± 0.031 bc	1.180 ± 0.020 a	112.005 ± 4.495 b
	T60	0.189 ± 0.004 a	0.988 ± 0.015 ab	1.203 ± 0.015 a	118.690 ± 3.118 a
	T80	0.197 ± 0.002 a	0.949 ± 0.020 bc	1.153 ± 0.012 b	109.747 ± 2.222 b
Flowering and fruit-setting stage	CK	0.183 ± 0.005 c	1.000 ± 0.028 c	0.997 ± 0.006 c	100.007 ± 2.962 c
	T0	0.195 ± 0.005 ab	1.061 ± 0.027 ab	1.317 ± 0.032 b	139.837 ± 3.547 b
	T20	0.194 ± 0.003 ab	1.058 ± 0.016 ab	1.347 ± 0.015 ab	142.400 ± 2.981 b
	T40	0.187 ± 0.006 bc	1.019 ± 0.031 bc	1.460 ± 0.062 a	148.84 ± 9.991 b
	T60	0.199 ± 0.002 a	1.084 ± 0.010 a	1.487 ± 0.050 a	160.880 ± 4.118 a
	T80	0.188 ± 0.008 bc	1.028 ± 0.042 bc	1.443 ± 0.015 a	148.470 ± 7.412 b
Fruiting stage	CK	0.161 ± 0.009 a	1.000 ± 0.057 a	0.997 ± 0.049 c	99.927 ± 5.851 c
	T0	0.128 ± 0.005 bc	0.797 ± 0.031 bc	1.227 ± 0.051 b	97.927 ± 5.580 c
	T20	0.133 ± 0.007 bc	0.825 ± 0.042 bc	1.430 ± 0.056 a	118.057 ± 10.320 abc
	T40	0.124 ± 0.002 c	0.768 ± 0.015 c	1.437 ± 0.119 a	110.607 ± 10.868 bc
	T60	0.131 ± 0.005 bc	0.877 ± 0.092 b	1.533 ± 0.061 a	135.047 ± 19.970 a
	T80	0.139 ± 0.005 b	0.866 ± 0.030 b	1.433 ± 0.081 a	123.85 ± 2.996 ab

Note: Different lowercase letters in the table represent significant differences (p < 0.05); using a single-factor method for analysis.

, during the seedling stage, L exhibited an upward trend with increasing organic fertilizer substitution rate. CK recorded 0.143, while T60 reached a peak of 0.165. T20–T80 significantly exceeded CK (p < 0.05). LI remained below 1 across all treatments, with T0 registering the lowest value (0.922) and T60 the highest (0.988). CPI and CPMI increased with the rising proportion of organic fertilizer replacing chemical fertilizer, with T40–T80 significantly higher than CK, and T60’s CPMI reaching 133.17. During the budding stage, CK had the highest L (0.194), T0 the lowest (0.179), and T60–T80 values approached CK; LI trends resembled those in the seedling stage, with T60 reaching 0.988; CPI and CPMI were significantly higher than CK during T60–T80, with T60’s CPI at 1.203 and CPMI at 118.69. During the flowering and fruit-setting stage, L reached 0.199 at T60, significantly higher than CK (0.183); LI increased with rising organic fertilizer substitution rates, reaching 1.084 at T60; CPI and CPMI were significantly higher than CK from T40 to T80, with T60 values of 1.487 and 160.88, respectively—approximately 49% and 61% higher than CK. During the fruiting and harvesting period, L values were lower than CK across all treatments, with T40 showing the lowest (0.124); LI values at T60–T80 were 0.877 and 0.866, higher than those at T0–T40. CPI and CPMI were significantly higher than CK at T20–T80, with CPI at T60 being 1.533 and CPMI at 135.047, representing increases of approximately 54% and 35% compared to CK.

3.4. Effect of Fertilization Practices on Yield and Cumulative CH₄ and CO₂ Emissions

As shown in Figure 4, the T60 treatment yielded the highest production, showing increases of 26.25%, 20.24%, 8.6%, 11.92%, and 6.32% compared to the CK, T0, T20, T40, and T80 treatments, respectively. The differences in yield among all treatments reached statistically significant levels. Cumulative CO₂ emissions under all fertilization treatments significantly exceeded those of the CK, with the T60 treatment exhibiting the highest cumulative CO₂ emissions. Compared to the CK, T0, T20, T40, and T80 treatments, the T60 treatment showed increases of 14.44%, 7.23%, 9.9%, 5.27%, and 6.43%, respectively. Except for no difference between the T0 and T80 groups, the cumulative CO₂ emissions between all other treatments reached statistically significant levels. CH₄ cumulative emissions were reduced under all fertilization treatments relative to the CK. Treatments T0, T40, T60, and T80 showed a significant decreasing trend, with T60 reaching the minimum value. This indicates that the net CH₄ uptake was highest under the T60 treatment.

By calculating the global warming potential (GWP) and greenhouse gas emission intensity (GHGI) for each treatment, it is evident that the trend in GWP changes and the level of intergroup differences largely align with cumulative CO₂ emissions. The GWP for all fertilization treatments was significantly higher than that of the CK. The T60 treatment exhibited the highest GWP, exceeding the CK, T0, T20, T40, and T80 treatments by 14.26%, 7.23%, 9.72%, 5.03%, and 6.33%, respectively. No significant differences were observed between T0 and T80, while the remaining treatments exhibited statistically significant differences in GWP when compared pairwise. Furthermore, the overall GWP levels corresponded to CO₂ and CH₄ emission levels. GHGI, derived as the ratio of GWP to yield, showed relatively balanced values across treatments.

3.5. Correlation Analysis of Soil Organic Carbon Fractions and Nutrients with Cumulative CH₄ and CO₂ Emissions

Figure 5 reveals the correlations among soil nutrients, carbon components, and cumulative emissions of CH₄ and CO₂ during the four growth stages of chili peppers.

(1) During the seedling stage, AK showed significant positive correlations (p < 0.05) with MOC, MBC, POC, and SOC, and positive correlations with EOC and DOC. AP exhibited positive correlations with carbon components and a significant positive correlation (p < 0.05) with EOC, indicating that soil potassium and phosphorus nutrients directly promote carbon component accumulation. Organic carbon components showed negative correlations with CO₂, highlighting the “carbon sequestration effect” where stable carbon pools suppress CO₂ release. Potassium fertilizer exhibited a negative correlation with methane, suggesting that potassium may inhibit methanogenic archaeal activity and reduce methane emissions.

(2) Correlations weakened during the budding stage, with stable carbon pool effects becoming more pronounced. Although positive correlations between nitrogen, phosphorus, potassium fertilizers, and carbon components persisted, their intensity (indicated by color concentration) significantly decreased compared to the seedling stage. This suggests heightened nutrient competition during the budding stage, with more nutrients transferred to the plant body, thereby weakening direct regulation of carbon pools. The relationship between EOC, MBC, and CO₂ shifted from negative to positive compared to the seedling stage. MOC exhibits a positive correlation with CO₂, reflecting enhanced “positive contribution” of the stable carbon pool (MOC) to CO₂ during the transition from “emission suppression” to “sequestration–emission equilibrium” in soil during the budding stage. AK maintains a negative correlation with CH₄ but shows a significant positive correlation with CO₂ (p < 0.05), indicating more complex pathways of nutrient regulation on gas emissions (simultaneously suppressing CH₄ emissions while contributing to CO₂ enhancement).

(3) During the flowering and fruit-setting stage, active carbon reactivated, intensifying competition. The positive correlation between NPK fertilizers and carbon components further weakened. Some indicators (AK and CH₄, TP and CO₂, and TN and CH₄) showed negative correlations. This occurred because vigorous reproductive growth during flowering and fruiting led to surges in root exudates (EOC sources) and microbial activity (MBC). Soil nutrients prioritized plant demands, indirectly affecting carbon–gas associations. As reproductive growth drives root exudates (active carbon) and microbial decomposition, the positive correlation between EOC, MBC, and CO₂ strengthens. MOC shows a highly significant positive correlation with CO₂ (p < 0.01), indicating that the aggregate structure becomes more critical for physical carbon protection during this stage. The correlation between TP and CH₄ shifts from negative to positive, reflecting a transition in CH₄ regulation from nutrient-driven mechanisms to plant–microbe interactions.

(4) During the yield stage, correlations between TP, TN, AK, AP content, carbon content, and gas emissions weakened, with no significant associations observed in most cases. By this point, the soil–crop system had matured, establishing a dynamic equilibrium among nutrients, carbon pools, and gas emissions. Direct fertilization effects were masked by systemic self-regulation. Although the positive correlation between organic carbon (EOC), microbial carbon (MBC), and CO₂ persists, its strength declines to the lowest point throughout the growth cycle. Microbial organic carbon (MOC) exhibits a highly significant positive correlation with CO₂ (p < 0.01), indicating that stable carbon pools become the primary carbon sink during the fruiting stage. The contribution of active carbon turnover to gas emissions decreases, and the system evolves toward a steady state characterized by “low emissions and high carbon sequestration”.

4. Discussion

4.1. Effect of Fertilization Practices on the Dynamics of Soil Carbon Pools and Their Indices During the Reproductive Period

In this study, the partial substitution of chemical fertilizers with vegetable-waste-based organic fertilizer significantly regulated the dynamics of the soil carbon pool, with effects varying across the growth stages of chili peppers.

At the seedling stage, the T60 treatment outperformed T0 in promoting mineral-associated organic carbon (MOC) and dissolved organic carbon (DOC), owing to enhanced formation of organo-mineral complexes facilitated by continuous organic carbon input (Figure 1). In contrast, the T0 treatment, which relied solely on chemical nutrients, resulted in less stable carbon accumulation due to limited microbial activity and root development. These findings align with Chen et al. (2025) [29] and Wei et al. 2025 [30], who reported that organic fertilization increases soil organic carbon (SOC), active fractions (e.g., MBC and DOC), and nutrient availability (e.g., TP and AP), while reducing soil bulk density and acidification. The significant positive correlation (p < 0.05) between available potassium (AK), total phosphorus (TP), and both MOC and EOC further supports the role of nutrient inputs in promoting the initial build-up of carbon pools, consistent with Sharma et al. (2022) [31].

By the budding stage, the direct stimulatory effect of organic fertilizer decreased but the sequestration capacity of stable carbon pools (e.g., MOC) became more pronounced, shifting the soil carbon economy toward a “sequestration–emission equilibrium”. As mentioned in the Introduction Section, the application of tail grass provides an easily accessible source of organic carbon, which we hypothesize will stimulate microbial activity and thereby influence greenhouse gas emissions. Our results indicate that organic fertilizers applied to tail grass can effectively suppress CO₂ emissions by enhancing the carbon sequestration effect of stable carbon pools. In contrast, the T0 treatment lacks organic carbon input, leading to low efficiency in constructing stable carbon pools and, consequently, poor carbon pool stability. These results are supported by Yang et al. (2023) [32] and Zhang et al. (2025) [33], who documented improved SOC content, microbial diversity, and carbon turnover balance under organic substitution.

During the flowering and fruit-setting stage, the positive correlation between stable carbon pools (e.g., MOC) and CO₂ emissions increased significantly (p < 0.01), reflecting that the “positive contribution” of stable carbon pools to CO₂ emissions increased with the increase in fertilizer requirement of the crop (Figure 5). Easily oxidizable organic carbon (EOC) and microbial biomass carbon (MBC) were positively correlated with CO₂ emissions, while the positive correlation between POC and CO₂ emissions weakened, indicating that organic fertilization simultaneously promoted active carbon turnover and physical protection through aggregation. Wang et al. (2025) [34] confirmed that long-term organic substitution significantly raised EOC, MBC, and POC contents and enhanced the proportion of stable C fractions, corroborating the dual role of emission and sequestration under organic management.

During the results period, the organic fertilizer treatment group maintained higher levels of MOC and POC while sustaining a stable positive correlation with carbon dioxide, confirming the establishment of a high carbon sequestration and low-emission carbon balance state. In contrast, the chemical fertilizer treatment group resulted in a carbon state characterized by high variability and low stability. Mon et al. (2024) [35] and Han et al. (2024) [36] also reported enhanced stable carbon pool content and sequestration efficiency under organic amendments, supporting the notion of long-term carbon homeostasis.

The soil carbon pool management index (CPMI) demonstrated an overall “increase-then-stabilize” trend with increasing organic fertilizer ratio (Table 3). At the seedling stage, T60 increased CPMI by 21.3% over T0, reflecting the co-accumulation of active and stable carbon pools driven by integrated nutrient–carbon inputs. Zhang et al. (2022) [37] and Lagomarsino et al. (2025) [38] reported similar mechanisms, where balanced nutrient supply enhanced CPMI through improved microbial activity and carbon storage. During the budding stage, the growth rate of CPMI slowed but the contribution of MOC to CPMI rose substantially, highlighting the role of stable carbon under organic regimes, consistent with Yan et al. (2025) [39]. At flowering and fruit-setting, CPMI peaked due to reactive carbon flux but was stabilized by POC-mediated protection, reducing CPMI fluctuation. Zhang et al. (2023) [40] observed similar buffering effects under combined organic–inorganic fertilization. By fruiting, CPMI variation narrowed under T60–T80, confirming better carbon quality–quantity balance, as also shown by He et al. (2023) [41] and Wang et al. (2024) [42].

In summary, substitution with vegetable-waste-based organic fertilizer enhanced the stability of soil carbon pools, promoted the formation of stable carbon components, optimized the carbon management index, and supported a transition toward low-emission, high-sequestration systems across chili pepper growth stages.

4.2. Effect of Fertilization Practices on Cumulative CO₂ and CH₄ Emissions and Warming Potential During the Reproductive Period

The application of organic fertilizer derived from vegetable waste reduced the global warming potential (GWP) over the entire growth cycle, exhibiting a stage-specific pattern of “early inhibition—mid-term synergism—late fixation” as the proportion of organic substitution increased.

At the seedling stage, CO₂ emissions were lower in the organic fertilizer replacement treatment group compared to the pure chemical fertilizer treatment. This reduction is attributed to the optimized microbial community structure and suppressed decomposition of active organic carbon due to the suitable carbon-to-nitrogen (C/N) ratio of the organic fertilizer. In contrast, the T0 treatment, relying solely on chemical nutrients, led to lower microbial carbon use efficiency and higher CO₂ emissions. Additionally, available potassium (AK) showed a significant negative correlation with CH₄ emissions (p < 0.05), indicating a suppressive effect on methane production. The GWP at this stage was dominated by CO₂, exhibiting a “high CO₂, low CH₄” emission pattern. These findings are consistent with Zhang et al. [37], who reported reduced CO₂ emissions under organic substitution due to microbial community optimization, and Lee et al., 2025 [43], who demonstrated AK-induced suppression of CH₄ in maize–kale rotations.

At the budding stage, CH₄ emissions from the organic fertilizer replacement treatments were reduced compared to the pure chemical fertilizer treatment. This phenomenon is primarily attributed to the altered carbon utilization pathways of microorganisms following the addition of vegetable residue organic fertilizer. The easily degradable organic carbon in vegetable residue (such as soluble sugars) preferentially stimulated the activity of aerobic microorganisms and certain anaerobic microorganisms (such as sulfate-reducing bacteria) in the soil. These microorganisms compete with methanogenic archaea for common substrates (such as H₂ and acetate), and, due to their higher competitive ability, they effectively inhibit the methanogenesis process, thereby reducing CH₄ emissions. Therefore, the addition of exogenous organic carbon does not directly promote methanogenesis but instead redirects carbon flow to other metabolic pathways through a stimulation effect. Although CO₂ remained the dominant contributor to GWP, overall emission intensity decreased, reflecting a transition of the carbon pool toward an equilibrium state. These results align with Mon et al., 2024 [35], where refractory carbon inputs and MOC sequestration jointly reduced GHG emissions.

During the flowering and fruit-setting stage, CO₂ emissions under T60 peaked due to reproductive growth; however, the sequestration effect of particulate organic carbon (POC) reduced GWP by 19.3% compared to T0. This underscores the role of physical protection by aggregates in offsetting carbon emissions. In the T0 treatment, the lack of effective sequestration mechanisms resulted in the highest GWP of the entire growth period, driven largely by increased root exudation and the positive correlations between EOC, MBC, and CO₂. Meanwhile, CH₄ emissions were regulated more by plant–microbe interactions (e.g., root oxygen secretion) than direct nutrient availability. Hu et al., 2023 [44] similarly reported lower GWP under organic fertilization linked to enhanced POC protection and reduced CO₂ and CH₄ emissions.

By the fruiting stage, T60–T80 treatments reduced GWP by 25.6–30.1% compared to T0. A negative correlation between stable carbon pools (MOC and POC) and CH₄ confirmed the establishment of an effective “emission–sequestration” balance under organic amendment. The T0 treatment still exhibited high GWP due to insufficient stable carbon pool formation. As the system stabilized, the turnover of active carbon slowed, the correlation with CO₂ weakened, and CH₄ remained low, leading to a “low emission, low warming” stage. These results are consistent with Pramono et al., 2022 [45] and Cui et al., 2025 [46], who reported enhanced stability of MOC and POC and significantly reduced GWP under integrated organic management.

Although organic substitution may stimulate early-stage emissions, it significantly strengthens carbon sequestration capacity in later stages, resulting in a long-term reduction in warming potential—demonstrating a “short-term stimulation, long-term regulation” effect on GWP.

4.3. Mechanisms of Soil Nutrient–Carbon Component–Greenhouse Gas Interactions During the Reproductive Period

In this study, we found that the regulatory mechanism of GHG emissions exhibited distinct stage-specific characteristics as crop fertility progressed. Initially, at the seedling stage, a “direct nutrient regulation of carbon–gas” mode was observed, whereby soil nutrients directly drove CO₂ and CH₄ emissions by influencing microbial activity—evidenced by a surge in CO₂ emissions and fluctuations in CH₄ under fertilizer treatment. As development advanced, this mode gradually shifted to a state of “weakening nutrient influence” by the fruiting and harvesting stage. During this later period, reproductive growth dominated nutrient allocation to vegetative organs, and the direct soil–nutrient–carbon–gas linkage was replaced by a “plant–soil” system competition. For example, the negative correlation between stable carbon pools (MOC and POC) and CO₂ was enhanced, and CH₄ emissions became regulated more by plant–microbial interactions (e.g., root oxygen secretion altering redox potential).

These findings align with Chen et al., 2025 [29], who reported in a full-life-cycle study of maize that chemical fertilizer stimulated microbial respiration via rapid inorganic nitrogen release at the seedling stage, increasing CO₂ emissions by 21.3% compared to organic fertilizer treatments, while CH₄ emissions showed strong nitrogen-driven fluctuations. By the fruit-harvesting stage, organic fertilizer treatments suppressed direct soil–gas emissions through enhanced root exudation of organic acids (increased by 34.6%) and improved MOC sequestration (up by 18.7%), reducing GWP by 27.4% compared to chemical fertilizer. This supports the conclusion that the fertility process drives a shift in regulatory mechanisms by reorganizing the “soil–plant” nutrient partitioning pattern during the reproductive period.

Across growth stages, the following occurred:

At the seedling stage, active carbon (EOC and MBC) dominated CO₂ emissions, while stable carbon pools (MOC) contributed to sequestration. Rapid decomposition of EOC led to increased CO₂ release.

At the budding stage, the sequestration effect of MOC was enhanced, slowing CO₂ emissions through physical protection mechanisms and potentially suppressing methanogenic activity, helping to establish an emission equilibrium.

During the flowering and fruit-setting stage, reproductive growth reactivated active carbon pools (EOC and MBC), strengthening their positive correlation with CO₂ and resulting in high emission rates.

By the fruiting stage, stabilized carbon pools (MOC and POC) dominated sequestration, significantly reducing GWP, while the contribution of active carbon diminished to a minimum.

This pattern is consistent with Aumtong et al., 2022 [47], who demonstrated that organic fertilization increased EOC and CO₂ emissions at the budding stage but enhanced MOC sequestration by the tillering stage (analogous to the budding stage), slowing CO₂ release. Furthermore, a positive correlation between POC and CO₂ at the filling stage (equivalent to fruiting) drove a significant reduction in GWP, supporting the role of carbon pool dynamics in stage-wise emission regulation.

This study also revealed that CH₄ emissions were initially inhibited by nutrients (e.g., available potassium), and later dynamically regulated through plant–microbe interactions (e.g., root oxygen secretion affecting redox potential). Meanwhile, CO₂ emissions were consistently linked to reactive carbon pools (EOC and MBC) but their driving intensity fluctuated across reproductive stages following an “enhanced (seedling stage) → weakened (budding stage) → re-enhanced (flowering and fruit-setting stage) → stabilized (fruiting stage)” pattern, synchronizing with crop carbon demand and soil carbon turnover rate.

Ansabayeva et al., 2025 [48] confirmed that PGPR inoculation suppressed methanogen activity via enhanced root organic acid secretion, reducing CH₄ emissions by 31.2%, while increased root oxygen secretion during the budding stage boosted CH₄ oxidation by 47.6%, aligning with the plant–microbe interactions observed herein. Similarly, Chi et al., 2025 [49] found in rice paddies that alternate wetting and drying (AWD) increased CO₂ emissions by 22.5% at the seedling stage compared to continuous flooding, and that peak CO₂ emissions at the budding stage (reaching 18.7 mg/kg/day) were driven by root respiration surge, whereas enhanced MOC sequestration during fruiting stabilized CO₂ emissions—consistent with the fertility-dependent fluctuation in CO₂ driving intensity found in this study.

Overall, this study demonstrates that the regulatory mechanisms of GHG emissions shift progressively during crop development—from direct nutrient-mediated control in early stages to plant–soil system interactions in reproductive phases. The dynamic interplay between active and stable carbon pools plays a critical role in modulating CO₂ emissions, while CH₄ emissions are increasingly influenced by plant-mediated microbial processes as fertility advances. These stage-specific mechanisms underscore the importance of tailoring management practices—such as nitrification inhibition during the budding stage or organic amendment during fruiting—to synchronize with physiological and carbon turnover dynamics for effective GHG mitigation and carbon sequestration.

5. Conclusions

This study revealed the regulatory mechanisms of combined application of chemical fertilizers and organic fertilizers on soil carbon pools and greenhouse gas emissions in tail vegetables during the growth period of peppers. The main conclusions are as follows:

(1) In terms of the dynamic regulation effect on carbon pools, the combined application of organic fertilizers (especially T60: 40% chemical fertilizer + 60% vegetable waste organic fertilizer) significantly increased soil organic carbon components (MOC, POC, etc.) at each growth stage. Among these, the retention effect of the stable carbon pool strengthened as the proportion of organic fertilizer increased. During the harvest period, MOC and POC contents were 18.7–22.4% higher than in the sole application of chemical fertilizers (T0) treatment. While the sole application of chemical fertilizers (T0) promoted short-term growth of active carbon (EOC and MBC), it failed to adequately construct stable carbon pools, and carbon pool stability was lower than in the combined application treatments.

(2) In terms of optimizing the carbon pool management index (CPMI), the soil CPMI under T60 treatment performed best throughout the entire growth cycle. Compared with T0, its CPMI increased by 19.3% from the seedling stage to 21.3% during the flowering and fruit-setting stage, reflecting the synchronous optimization of carbon pool quality and quantity. The combined application of high-proportion organic fertilizers (T60–T80) enhanced the stabilizing role of the carbon pool, maintaining the CPMI at a high and stable level during the fruiting stage. Compared to T0, the fluctuation range was reduced by over 50%.

(3) Greenhouse gas mitigation effects: The treatments significantly reduced CO₂ and CH₄ emissions and greenhouse potential (GWP), with GWP reduced by 25.6% at T60 compared to T0, and peak CO₂ emissions at flowering and fruit-setting by 19.3% due to POC sequestration. Organic fertilizer for tail grass achieves the synergistic effect of “phase reduction–long term sequestration” by optimizing the microbial community (at the seedling stage), suppressing methanogenic archaea (at the budding stage), and enhancing carbon sequestration (at the fruiting stage).

(4) Optimal fertilization patterns and their practical significance: The optimal pattern (T60) involves the combined application of 40% chemical fertilizer and 60% organic fertilizer. This pattern balances the carbon cycle and carbon sequestration processes, achieving the dual benefits of increasing carbon pool capacity and reducing emissions. It is recommended to dynamically adjust fertilization methods according to production stages: establish the carbon pool foundation during the seedling stage, regulate carbon cycling during the budding stage and flowering and fruit-setting stage, and strengthen and stabilize the carbon pool during the fruiting stage. This approach can reduce the global warming potential (GWP) by approximately 20% to 30%.

This study has several important limitations that highlight the context-specific nature of its findings. First, the experimental site is located in a semi-arid region characterized by typical loess (yellow loess soil), with unique climatic conditions (such as limited rainfall and high evaporation) and soil properties (such as alkaline pH and moderate organic matter content). Therefore, the conclusions may not be directly applicable to regions with markedly different environmental conditions (e.g., humid or semi-humid zones) or areas with distinctly different soil properties (e.g., acidic soils, clay soils) as carbon turnover dynamics and greenhouse gas emissions may vary significantly in these regions.

Second, this study focused exclusively on organic fertilizers produced from vegetable waste (vegetable trimmings). It remains unclear whether the optimal substitution ratio identified (e.g., T60: 40% chemical fertilizer + 60% organic fertilizer) applies to other organic amendments (e.g., livestock manure, crop straw, or composted municipal solid waste). These materials exhibit significant differences in nutrient release patterns, microbial degradability, and organic carbon composition, which may influence their effects on soil carbon sequestration and greenhouse gas emissions. Therefore, experimental validation is required to determine optimal substitution ratios when using other organic wastes.

Finally, this study specifically addresses chili peppers (Longjiao No. 2) cultivated under protected agriculture conditions. Soil carbon pool responses and greenhouse gas emissions may vary across crop types due to differences in root exudates, nutrient uptake patterns, and microbial interactions. Thus, caution is warranted when extrapolating findings to other cropping systems or plant species.

Future research should prioritize multi-site trials across diverse agroecological zones and soil types to evaluate various organic amendments, incorporating different crops to develop more comprehensive and universally applicable sustainable fertilization management recommendations.