Cereal-Legume Mixed Residue Addition Increases Yield and Reduces Soil Greenhouse Gas Emissions from Fertilized Winter Wheat in the North China Plain

Abstract

Incorporating crop residues into the soil is an effective method for improving soil carbon sequestration, fertility, and crop productivity. Such potential benefits, however, may be offset if residue addition leads to a substantial increase in soil greenhouse gas (GHG) emissions. This study aimed to quantify the effect of different crop residues with varying C/N ratios and different nitrogen (N) fertilizers on GHG emissions, yield, and yield-scaled emissions (GHGI) in winter wheat. The field experiment was conducted during the 2018–2019 winter wheat season, comprising of four residue treatments (no residue, maize residue, soybean residue, and maize-soybean mixed residue) and four fertilizer treatments (control, urea, manure, and manure + urea). The experiment followed a randomized split-plot design, with N treatments as the main plot factor and crop residue treatments as the sub-plot factor. Except for the control, all N treatments received 150 kg N ha⁻¹ season⁻¹. The results showed that soils from all treatments acted as a net source of N₂O and CO₂ fluxes but as a net sink of CH₄ fluxes. Soybean residue significantly increased soil N₂O emissions, while mixed residue had the lowest N₂O emissions among the three residues. However, all residue amendments significantly increased soil CO₂ emissions. Furthermore, soybean and mixed residues significantly increased grain yield by 24% and 21%, respectively, compared to no residue amendment. Both soybean and mixed residues reduced GHGI by 25% compared to maize residue. Additionally, the urea and manure + urea treatments exhibited higher N₂O emissions among the N treatments, but they contributed to significantly higher grain yields and resulted in lower GHGI. Moreover, crop residue incorporation significantly altered soil N dynamics. In soybean residue-amended soil, both NH₄⁺ and NO₃⁻ concentrations were significantly higher (p < 0.05). Conversely, soil NO₃⁻ content was notably lower in the maize-soybean mixed residue amendment. Overall, our findings contribute to a comprehensive understanding of how different residue additions from different cropping systems influence soil N dynamics and GHG emissions, offering valuable insights into effective agroecosystems management for long-term food security and soil sustainability while mitigating GHG emissions.

1. Introduction

Global agriculture is confronted with the dual challenge of increasing crop productivity and sustainability to meet growing demands for food production while mitigating environmental impacts. Crop residue management, particularly the incorporation of residues from preceding crops into agricultural soils, has been widely accepted as a way to improve soil fertility and crop yield [1], increase soil carbon (C) sequestration, and offset greenhouse gas (GHG) emissions in agriculture [2]. The carbon stored in residues can either be retained in the soil as organic matter or released back into the atmosphere through microbial decomposition [3], affecting the overall carbon budget of agricultural ecosystems. It has been estimated that global soil C sequestration could reach approximately 0.3 billion tons per year with crop residue amendment [4]. This potential of soil C sequestration, however, can be offset if crop residue amendment substantially increases soil GHG emissions, particularly nitrous oxide (N₂O), carbon dioxide (CO₂), and methane (CH₄).

China, as the world’s largest crop residue producer, contributes approximately 20% of the total global production, with an estimated national crop residue production of 737 × 10⁶ tons [5]. Traditionally, crop residues were often burned or left in/around the fields. Currently, around 30% of total crop residues in China are being incorporated into the soil [6,7]. The Chinese government has also implemented subsidies and supportive policies to encourage farmers to adopt residue incorporation into the soil in recent years [6]. Consequently, managing these crop residues has become a significant concern for scientists and policymakers [8].

Several studies have reported that incorporating crop residues into the soil substantially enhances soil microbial biomass and activity, accelerates organic matter decomposition rates, and promotes nutrient cycling processes [9,10], resulting in a significant increase in CO₂ and N₂O emissions from the soil [11]. Moreover, crop residue incorporation can induce changes in various soil parameters, including soil mineral N (NH₄⁺ and NO₃⁻) concentration, dissolved organic carbon (DOC), pH levels, C/N ratio, soil moisture, and temperature, which collectively dictate the magnitude and direction of these emissions. The rate of CO₂ and N₂O emissions after residue incorporation primarily depends on the quality and quantity of residues [10,12,13]. Among residue characteristics, the most important properties are the C and N concentrations of the residues as well as the C/N ratio, which determines the dynamics of organic N and C in the soil [14]. A meta-analysis of existing literature also revealed that the emission factor (EF_N2O) of crop residues varies widely among cereal and legume residues due to their varying C/N ratios [4]. Notably, crop residues with low C/N ratios, such as legumes, tend to result in higher N₂O emissions due to rapid mineralization [15,16,17]. Conversely, residues with high C/N ratios may lead to net N immobilization during decomposition, thereby reducing N₂O emissions [4,18].

The dominant biological processes contributing to N₂O emissions include nitrification, denitrification, and nitrifier denitrification [19]. Specific soil physicochemical properties also play a crucial role in enhancing microbial formation of N₂O. For instance, the presence of soil mineral N (NH₄⁺ and NO₃⁻) and a carbon (C) source in the soil create favorable conditions for N₂O production through nitrification and denitrification processes [11,20]. Additionally, the introduction of a labile C source, such as cereal crop residue or root exudates, can stimulate microbial activities, leading to increased CO₂ emissions. Crop residues serve as organic C substrates for microbial growth, thereby promoting microbial N assimilation. Consequently, strong competition for NH₄⁺ between autotrophic nitrifiers and heterotrophic microorganisms results in N immobilization and reduced N₂O production [12,21]. Furthermore, crop residues provide energy for denitrifiers, thereby enhancing denitrification and N₂O production under anaerobic conditions [4]. Overall, soil moisture significantly influences nitrification and denitrification activities by regulating soil oxygen availability and redox potential [22]. It has been observed that nitrification dominates soil N₂O production when the soil has 30–60% water-filled pore space (WFPS), while denitrification rates rapidly increase when the WFPS exceeds 60% [4,22].

Among the crop residues, cereal (e.g., maize) and legume (e.g., soybean) residues, varying in biochemical composition, lignocellulosic content, and decomposition rates, exert significant influence on soil physicochemical properties, microbial dynamics, nutrient cycling, and subsequent soil GHG emissions. Maize residues are typically characterized by higher lignocellulosic content, a higher C/N ratio, and slower decomposition rates [23]. Conversely, soybean residues, with relatively lower lignin content, higher N concentration, and a lower C/N ratio, tend to decompose faster, offering immediate nutrient availability, resulting in higher nitrification and denitrification in the soil compared to maize residues [24]. The incorporation of a mixed residue involving maize and soybean residues, often observed in intercropping or sequential cropping systems, represents a confluence of contrasting residue characteristics. The synergistic effects of carbon-rich maize and nitrogen-rich soybean residues could alter soil microbial activity, C and N cycling, and nutrient availability [9], potentially affecting crop growth, productivity and soil GHG emissions differently than single-crop residue incorporation practices. While the effect of cereal and legume residue retention on soil nutrient dynamics and GHG emissions has been extensively studied before [3,12,13,16], the impact of cereal-legume mixed residue, especially maize-soybean mixed residue amendment, on soil nutrient dynamics and GHG emissions remains unknown. Thus, understanding the specific impacts of maize, soybean, and their mixed residue incorporation on soil GHG emissions is essential for devising sustainable agricultural practices that minimize environmental impacts while maintaining soil fertility and productivity.

Another crucial agricultural practice, alongside crop residue incorporation, is the application of N fertilizer. N fertilizers are commonly applied to foster straw mineralization and ensure sufficient available N for crop growth. However, N addition not only promotes residue mineralization and crop yield but also significantly increases soil GHG emissions [25,26]. Urea, due to its rapid denitrification in the soil, has the highest potential for field-scale GHG emissions among synthetic N fertilizers [27]. Mairura et al. [28] reported that combined applications of manure and synthetic fertilizers can substantially reduce global warming potential and yield-scaled GHG emissions without compromising yield. Based on this evidence, we can hypothesize that substituting a part of urea with manure could effectively reduce GHG emissions from agricultural fields.

While the effects of residue incorporation on soil N₂O emissions have been extensively studied [12,29,30], the net GHG emissions, including N₂O, CO₂, and CH₄, from crop residue, particularly from maize-soybean mixed residue amendment, have not been thoroughly investigated. Furthermore, the synergistic effects of crop residue amendment and types of N fertilizer on soil GHG emissions from winter wheat are unknown and need further research. The objective of the present study was (1) to investigate the effects of incorporating different crop residues on N₂O, CO₂, and CH₄ emissions, as well as global warming potential, under different types of N fertilizer treatments; (2) to investigate the effect of crop residue and N fertilizer on winter wheat yield and yield-scaled GHG emissions (GHGI); and (3) to analyze the effect of crop residue amendment and N fertilizer type on various soil properties and their relationship with N₂O, CO₂, and CH₄ emissions from the soil.

2. Materials and Methods

2.1. Site Description

The field experiment spanned the 2018–2019 winter wheat growing season and was conducted at the Luancheng Agro-Ecosystem Experimental Station (37°89′ N, 114°68′ E; elevation 50 m above sea level), part of the Chinese Academy of Sciences in the North China Plain (NCP) region of Hebei province. Initially, our field experiment was designed to span two growing seasons (2018–2019 and 2019–2020). However, due to the lockdown during the COVID-19 pandemic in China, data collection was halted for the second half of the 2019–2020 growing season. This NCP region features a temperate, semi-arid monsoon climate typified by cold winters and scorching summers. The annual average temperature in 2018 and 2019 was 14 °C, while the annual average precipitation was 368 mm, with the majority falling in July, August, and September. The average, maximum, and minimum air temperatures during the growing season were 8 °C, 38 °C, and −17 °C, respectively. Figure 1 depicts the daily mean, maximum, and minimum air temperatures, alongside precipitation levels during the experimental period. All meteorological data were obtained from the Luancheng Meteorological Station, which is located within the Luancheng Experimental Station. The soil in this location is classified as a silt loam Haplic Cambisol [31]. Within the top 0–20 cm soil layer, the soil contains 15 g kg⁻¹ of organic matter, 1.1 g kg⁻¹ of total nitrogen, 15 mg kg⁻¹ of available phosphorus (P-Olsen), and 95 mg kg⁻¹ of exchangeable potassium.

The NCP serves as a pivotal grain-producing region in China, contributing 75% and 35% to the nation’s wheat and maize production, respectively [32]. The prevalent cropping system in this region is the winter wheat-summer maize (or soybean) double-cropping system. Both wheat and maize are irrigated via flooding irrigation using groundwater. Summer maize (or soybean) is planted around mid-June and harvested in early October, while winter wheat is sown in mid-October and harvested in early June.

2.2. Experimental Design

The field experiment was initiated in October 2018, encompassing four crop residue and four nitrogen (N) treatments. The crop residue treatments included (i) no residue (R0), (ii) maize residue (MR), (iii) soybean residue (SR), and (iv) maize-soybean mixed residue (MSR). Concurrently, the N treatments encompassed (i) control (no nitrogen), (ii) urea, (iii) manure, and (iv) a combination of manure and urea. The residues used in this experiment were obtained from a maize-soybean intercrop study conducted in the respective plot during the summer season, employing the same N treatments. This means that the maize-soybean mixed residue comes from the maize-soybean intercrop, while the maize and soybean sole residues come from their respective monocultures. On average, maize residue amounted to 9.1 t ha⁻¹, soybean residue amounted to 3.04 t ha⁻¹, and maize-soybean mixed residue amounted to 8.2 t ha⁻¹ (with maize residue contributing 7.1 t ha⁻¹ and soybean residue contributing 1.1 t ha⁻¹) were incorporated into the soil. The average C/N ratios of maize residue, soybean residue, and maize-soybean mixed residue were 62, 16, and 46, respectively. In plots designated for the no residue treatment, maize cultivation took place during the summer season, and residues were manually removed post-harvest. Subsequently, all crop residues were shredded into small pieces (<10 cm fragments) using a mechanical shredder.

All N treatments, excluding the control, received 150 kg N ha⁻¹ season⁻¹, applied either entirely as a basal application or split equally into two applications. Details regarding the N, P, and K applications are presented in Table S1. The composted poultry manure utilized (obtained from Shijiazhuang Ikos Agricultural Technology Co., Ltd., Shijiazhuang, China) contained 40% organic matter, 0.7% mineral N, 2.1% total N, 1.9% P₂O₅, and 1.1% K₂O. The manure treatment received the full N dose as a basal application, while the urea and manure + urea treatments received half of the N (75 kg N ha⁻¹) as a basal application, with the remaining half applied as top dressing. Additionally, all treatments received 75 kg P₂O₅ ha⁻¹ as calcium superphosphate and 100 kg K₂O ha⁻¹ as potassium chloride before planting. Manure and fertilizer along with crop residues were uniformly incorporated into the soil to a depth of 15 cm using a rototiller within 3 h of application.

The experiment followed a randomized split-plot design, where the main plot factor consisted of the N treatment, and the subplot factor encompassed the crop residue treatment, each replicated three times. Each sub-plot was 160 m² (16 m × 10 m) in size and was separated by a 1 m buffer. Winter wheat (Triticum aestivum cv. Shiluan02-1) was planted in a north-south orientation, with rows spaced 15 cm apart and a seeding rate of 170 kg ha⁻¹. To ensure adequate soil moisture for seed germination, one irrigation was administered one week before planting, complementing the three irrigation cycles during the growing season. The management regimen included herbicide and insecticide applications, weed control, and other standard agricultural practices aligned with local farming norms. Further details regarding management activities are available in Table S2. Harvesting of the wheat crop took place using mechanical harvesters on 9 June 2019. Subsequently, all crop residues were reintegrated into the plots, receiving identical N treatments as before, and incorporated into the soil using a rototiller to prepare for the subsequent summer crop.

2.3. Gas Flux Measurements

The gas samples were collected using non-steady-state static chambers. Immediately after sowing, in each plot, an open-ended base collar made of polymethyl methacrylate (60 cm length, 20 cm width, and 7 cm height) was placed into the soil at a depth of approximately 5 cm. To minimize edge effects, these collars were positioned 3–4 m inside the plot and remained in place from planting to harvesting. The gas sampling chambers, also constructed from temperature-isolated polymethyl methacrylate (PMMA) measuring 60 cm × 20 cm × 40 cm (length × width × height), were equipped with an internal fan powered by a battery. This fan ensured uniform gas concentration and air temperature within the chamber. Additionally, each chamber featured a built-in thermometer to monitor internal temperature and a sampling tube with a three-way stopcock, securely attached to prevent any leakage.

Gas sampling occurred in the morning between 8 a.m. and 11 a.m., a period often considered representative of average daily flux emissions [33]. Sampling frequency varied throughout the cropping season: before and after winter, it was weekly; during peak winter (December–February), it was monthly due to low emissions resulting from frost conditions in the topsoil. However, after fertilizer application, gas samples were collected twice a week for two weeks, totaling 22 sampling events during the cropping season. During each sampling event, the chamber was positioned on the base collar for 60 min, and gas samples were drawn at 0, 20, 40, and 60 min intervals via the three-way stopcock. A water seal was employed between the base collar and chamber to prevent gas leakage during sampling. Each collection involved drawing 60 mL of gas using a polypropylene syringe connected to stopcocks and transferring it into pre-evacuated 100 mL gas sampling bags (Delin, Dalian, China) with a 25-gauge needle. Two additional thermometers were placed within the chamber during gas sampling, one at ground level and another at a 5 cm soil depth, to record ground and soil temperatures. To avoid potential GHG contributions from plant respiration, any weeds within the chamber base were clipped and removed before each sampling event.

The concentrations of N₂O, CO₂, and CH₄ were quantified using gas chromatography (Agilent GC-6820, Agilent Technologies Inc., Santa Clara, CA, USA) equipped with a ⁶³Ni Electron Capture Detector (ECD), a Flame Ionization Detector (FID), and a Thermal Conductivity Detector (TCD) for the detection of N₂O, CO₂, and CH₄, respectively. Calibration was performed using standard gases. Soil GHG fluxes were calculated as the rate of change in gas concentration within the chamber headspace over the 60 min collection period. Gas flux rate F (µg m⁻² h⁻¹) was calculated using the following formula:

$$F=\frac{dc}{dt}\times \frac{M}{V_0}\times \frac{V}{A}\times \frac{273}{273+T}\times \frac{P}{P_0}\times 60$$

where dc/dt is the slope of the changes of gas concentration over time in the chamber (ppbv min⁻¹); M is the relative molecular mass of N₂O (44 g mol⁻¹), CO₂ (44 g mol⁻¹), and CH₄ (16 g mol⁻¹); V₀ is the volume of an ideal gas (22.41 g mol⁻¹); V is the volume of the chamber (m³); A is the soil surface area occupied by the chamber base (m²); T is the temperature (°C) inside the chamber; P is the atmospheric pressure (hPa) during gas sampling; P₀ is the standard atmospheric pressure (hPa); and 60 is the conversion factor for minutes to hour.

Seasonal cumulative emissions of N₂O, CO₂, and CH₄ (kg ha⁻¹ season⁻¹) from planting to harvesting were estimated by linear interpolation between successive sampling days, as described by Zhai et al. [34]:

$$Seasonalemissions=\sum \left\{\frac{\left(F_{i+1}+F_i\right)}{2}\times \left(t_{i+1}-t_i\right)\right\}\times 24\times \frac{1}{\mathrm{100,000}}$$

where F_i and F_i+1 are the fluxes of N₂O, CO₂, and CH₄ (µg m⁻² h⁻¹) at the previous and current gas sampling dates; t_i and t_i+1 are the previous and the current gas sampling dates; 24 is used to convert fluxes from h⁻¹ to d⁻¹; and 1/100,000 is used to convert fluxes from µg m⁻² to kg ha⁻¹.

The net global warming potential (Kg CO₂-eq ha⁻¹ season⁻¹) was calculated using the warming potential coefficient (CO₂ equivalent) of 298 for N₂O, 34 for CH₄ and 1 for soil CO₂ emissions, based on a 100-year time scale of IPCC AR6 [35].

Yield-scaled GHG emissions (GHGI) were calculated using the following formula:

$$GHGI=\frac{GWP}{GY}$$

where GWP is the global warming potential (Kg CO₂-eq ha⁻¹ season⁻¹), and GY is the grain yield (kg ha⁻¹).

2.4. Soil and Plant Sampling and Analysis

2.4.1. Soil Sampling

Soil samples were collected at two-week intervals from the furrow slice depth of 0–15 cm, coinciding with the gas sampling schedule. The top 0–15 cm depth of soil was chosen as the focus of our study, as it encompasses near-surface soil processes that significantly influence nutrient dynamics and GHG emissions. Previous studies have indicated that crucial soil processes often occur within the 0–10 or 0–15 cm soil depths, which play a substantial role in soil nutrient dynamics and GHG emissions [36,37]. However, no soil sampling occurred during winter (from December to February) due to frost conditions of the soil surface. Each sample, obtained using a 4 cm diameter augur, represented an amalgamation from 4–5 random locations within each plot. Post collection, the soils from each plot were thoroughly mixed and refrigerated at 4 °C until further analysis.

2.4.2. Analysis of Soil Physical and Chemical Properties

The determination of gravimetric soil moisture content involved oven drying 60–70 g of field-moist soil at 105 °C for 48 h. Soil NO₃⁻ and exchangeable NH₄⁺ content were evaluated by extracting 10 g of fresh soil with 50 mL of 1 M KCl solution (using a 1:5 w/v extraction ratio), followed by agitation for 60 min and filtration through Q5 filter paper, as previously described by Liu et al. [38]. Soil NO₃⁻ content was assessed using a dual-wavelength ultraviolet spectrophotometer (UV-2450, Shimadzu Corporation, Kyoto, Japan). The measurement of soil exchangeable NH₄⁺ content was conducted using the Smartchem 140 discrete chemistry analyzer (AMS Alliance, Frepillon, France).

For the determination of soil C/N ratio, air-dried soil was ground and sieved through a 0.2 mm sieve. Subsequently, 150 mg of soil was used to prepare a capsule with tin foil and analyzed via dry combustion using a macro elemental analyzer (vario MACRO cube, Elementar, Hanau, Germany). Soil pH was assessed in a suspension containing 10 g of dry soil and 25 mL of distilled water (at a 1:2.5 w/v ratio) following 30 min of shaking. Analysis of soil dissolved organic carbon (DOC) involved extracting 10 g of soil with 50 mL of distilled water (using a 1:5 w/v ratio) and shaking for 60 min. The resulting solution underwent filtration through Q5 filter paper, followed by centrifugation at 8000 rpm for 5 min. The supernatant was further filtered through a 0.45 µm filter membrane and analyzed using a vario TOC cube auto-analyzer (Elementar, Hanau, Germany).

2.4.3. Yield and Biomass Analysis

To assess the impact of crop residue and N treatments on biomass production at different growth stages, above-ground plant biomass was collected from 0.45 m² (3 rows for 1 m) at five distinct growth stages. The specific sampling dates were 24 March, 13 April, 29 April, 12 May, and 7 June, corresponding to the seedling, jointing, heading and booting, flowering, and maturity stages, respectively. Following collection, the plant samples were oven-dried at 65 °C until reaching a constant weight to determine the dry weight.

During physiological maturity, wheat samples were manually harvested from a 4 m² area (2 m × 2 m) in each plot. The samples were then threshed using a mechanical thresher, sun-dried to achieve the standard 13% moisture content, and subsequently weighed to determine the grain yield.

2.5. Statistical Analysis

Data for all parameters were subjected to a normality assessment using the Shapiro–Wilk test, revealing non-normal distributions (p < 0.05), even following data transformation attempts. The impact of cropping systems and N treatments on various parameters (cumulative seasonal N₂O, CO₂, and CH₄ emissions; grain yield; net global warming potential; and yield-scaled emissions) was assessed using analysis of variance (ANOVA). ANOVA was calculated using a general linear model, and when the main effects were significant (p ≤ 0.05), pairwise comparisons were analyzed using a post hoc least significant difference (LSD) test. To analyze the effect of cropping systems and N treatments on soil properties and environmental factors, data were analyzed by a non-parametric Kruskal–Wallis test. Significant main effects (p ≤ 0.05) underwent post hoc Mann–Whitney tests for pairwise comparisons.

The relationships between GHG fluxes (N₂O, CO₂, and CH₄), soil properties (NH₄⁺, NO₃⁻, soil C/N ratio, pH, and DOC concentrations), and environmental factors (soil moisture content and ground temperature, soil temperature at 5 cm depth) were performed using Spearman’s rho rank correlation coefficient analysis (p < 0.05; two-tailed test). ANOVA and non-parametric tests were performed by Minitab^® 17 Statistical software (Minitab Inc., State College, PA, USA), while correlation analysis was performed by OriginPro 2021 software (OriginLab, Northampton, MA, USA).

3. Results

3.1. Weather Conditions

During the winter period (from December to February), the topsoil was in frost condition, with average air and soil (5 cm depth) temperatures of −1.5 °C, and −0.5 °C, respectively. The total precipitation during the growing season was 88 mm, significantly below the crop water requirements. To address moisture deficits in the soil, a series of three irrigations were administered in each plot (160 m²), totaling 60 mm, 65 mm, and 70 mm. As the growing season progressed, particularly during the flowering and maturity stages, daytime temperatures increased noticeably (Figure 1), reaching an average daily temperature of 23 °C.

3.2. Grain Yield and Biomass Dry Matter Production

Our ANOVA results showed that the incorporation of crop residues and nitrogen fertilizers, as well as their interactions, significantly affected winter wheat grain yield (

Table 1. ANOVA (Pr > F) of N₂O, CO₂, and CH₄ emissions; global warming potential; grain yield; and yield-scaled emissions (GHGI) as affected by crop residue incorporation and N treatments.

Variable Source	d.f.	N2O Emissions	CO2 Emissions	CH4 Emissions	Global Warming Potential	Grain Yield	GHGI
Crop residue	3	0.031	0.001	0.891	0.002	0.038	0.031
N treatment	3	0.000	0.636	0.796	0.368	0.000	0.000
Crop residue × N treatment	9	0.046	0.006	0.924	0.030	0.008	0.048

d.f. is degree of freedom.

). Both soybean and maize-soybean mixed residue amendments resulted in significantly higher yields (p < 0.05) compared to no residue, exhibiting increases of 24% and 21%, respectively (

Table 2. Seasonal N₂O, CO₂, and CH₄ emissions (expressed as kg ha⁻¹ season⁻¹); grain yield (kg ha⁻¹), global warming potential (GWP) (expressed as Kg CO₂-eq ha⁻¹ season⁻¹); and yield-scaled emissions (GHGI) (Kg CO₂-eq kg⁻¹ grain) from different crop residue amendments and N treatments during study period at the Luancheng Agro-Ecosystem Experimental Station.

Treatment	N2O Emissions	CO2 Emissions	CH4 Emissions	GWP (kg CO2-eq ha−1 season−1)	Grain Yield (kg ha−1)	GHGI (kg CO2-eq kg−1 grain)
	kg ha−1 season−1
R0	1.55 ± 0.26 b	6032 ± 214 b	−0.76 ± 0.08 a	6468 ± 264 b	4003 ± 470 b	1.81 ± 0.16 b
MR	1.69 ± 0.30 b	7640 ± 335 a	−0.80 ± 0.10 a	8116 ± 336 a	4231 ± 504 ab	2.29 ± 0.30 a
MSR	1.63 ± 0.25 b	6969 ± 249 a	−0.88 ± 0.12 a	7424 ± 211 a	4843 ± 449 a	1.71 ± 0.19 b
SR	2.11 ± 0.28 a	6884 ± 263 a	−0.87 ± 0.15 a	7484 ± 278 a	4946 ± 425 a	1.70 ± 0.23 b
Control	0.75 ± 0.10 c	6751 ± 322 a	−0.83 ± 0.13 a	6947 ± 337 a	2818 ± 306 c	2.71 ± 0.27 a
Urea	2.45 ± 0.22 a	6905 ± 252 a	−0.73 ± 0.11 a	7612 ± 281 a	5565 ± 352 a	1.43 ± 0.10 c
Manure	1.43 ± 0.11 b	7152 ± 313 a	−0.89 ± 0.12 a	7547 ± 312 a	3942 ± 249 b	1.96 ± 0.12 b
Manure + urea	2.35 ± 0.26 a	6717 ± 332 a	−0.88 ± 0.09 a	7387 ± 315 a	5699 ± 340 a	1.34 ± 0.12 c

Values are the means (n = 12) ± standard error (SE). Means with different lowercase letters in a column indicate significant differences (p < 0.05) among the treatments.

). Although not statistically significant, soybean and mixed residue amendments yielded 17% and 14% more grain, respectively, compared to maize residue amendment. Correspondingly, soybean and mixed residue treatments also notably impacted biomass production at various growth stages (Figure 2a). By the maturity stage, both treatments produced significantly higher biomass than the no-residue treatment. Additionally, incorporation of maize residue showed a significantly lower impact on wheat biomass production compared to the soybean residue treatment.

Regarding the nitrogen treatments, wheat yield showed significant increases in the urea and manure + urea treatments in contrast to the manure and control (Figure 2b). Consistently, throughout the growing season, urea and manure + urea amendments demonstrated a robust impact on biomass yield, culminating in significantly higher biomass at the maturity stage compared to the manure and control treatments.

3.3. Crop Residue Effect on Soil Mineral N Concentration

The dynamics of mineral N (NH₄⁺ and NO₃⁻) concentration in the soil are illustrated in Figure 3. The effect of crop residue on soil mineral N concentration varied throughout the growing season. Following the incorporation of crop residues, the NH₄⁺ concentration in the soil was relatively lower until the end of winter, then gradually increased until the heading and booting stage, and subsequently decreased until harvesting. In contrast, the NO₃⁻ concentration in the soil was higher during the first half of the growing season, gradually declining until the flowering stage, and then increasing again until harvesting. Crop residue incorporation also showed a notable effect on soil mineral N concentration across different treatments. Our results indicate that NH₄⁺ and NO₃⁻ concentrations were higher in the soils with soybean residue amendment during the initial stages of the growing season, suggesting rapid mineralization of soybean residue. Conversely, the NH₄⁺ concentration in the no-residue treatment was consistently lower in most soil sampling events.

3.4. Soil N₂O Emissions

Our ANOVA results indicate that the incorporation of crop residues, N fertilizers, and their interactions significantly influenced seasonal N₂O emissions from the soil (Table 1). However, N₂O fluxes from residue amended soils and N treatments varied across the gas sampling events. Throughout the growing season, N₂O fluxes ranged from a minimum value of 2.5 µg m⁻² h⁻¹ to a maximum value of 600 µg m⁻² h⁻¹ in crop residue treatments, while in N treatments, fluxes ranged from −2 to 950 µg m⁻² h⁻¹ (Figure 4). Notably, N₂O fluxes in the soybean residue amendment were consistently higher in most daily flux measurements, reaching a peak of 600.3 µg m⁻² h⁻¹ after top dressing. Consequently, soybean residue exhibited higher N₂O emissions under all N treatments (Figure 5) and significantly increased cumulative N₂O emissions by 36%, 27%, and 29% compared to no residue, maize residue, and maize-soybean mixed residue amendment, respectively (Table 2). However, total N₂O emission from maize residue and maize-soybean mixed residue did not significantly differ from that of the no-residue treatment.

Additionally, N fertilizers significantly affected soil N₂O emissions. Peak N₂O fluxes occurred shortly after nitrogen fertilizer application during basal application and top dressing, coinciding with rainfall or irrigation events, and lasted for 1–2 weeks before returning to background emissions (Figure 4). Notably, the highest emissions occurred during top dressing rather than basal application of fertilizer. Consequently, both urea and manure + urea treatments produced significantly higher cumulative N₂O emissions, which were 71% and 64% higher than manure amended soil, respectively, despite the same amount of N applied.

Furthermore, our correlation analysis revealed that soil N₂O fluxes had a significant positive correlation with soil moisture content (r = 0.45; p < 0.01) (Figure 6), supporting our observations of higher N₂O fluxes during irrigation and rainfall events. N₂O fluxes also showed a significant positive correlation with soil temperature (r = 0.32; p < 0.01). Throughout the growing season, N₂O emissions were positively correlated with soil NH₄⁺ (r = 0.41; p < 0.01) and NO₃⁻ contents (r = 0.36; p < 0.01) but negatively associated with soil pH (r = −0.3; p < 0.05).

3.5. Soil CO₂ Emissions

Crop residue amendments and their interactions with N treatments had a significant effect on soil CO₂ emissions (p < 0.01) (Table 1). Following the incorporation of crop residues, soil CO₂ fluxes were initially high but gradually decreased thereafter. These fluxes remained lower during the winter season and returned to their initial levels during the spring season (Figure 7). In most gas sampling events, maize residue exhibited higher soil CO₂ fluxes compared to other residue treatments (Figure 7). However, all crop residue amendments significantly increased cumulative soil CO₂ emissions (Table 2). Compared to no-residue amendment, maize residue, soybean residue, and maize-soybean mixed residue amendment led to a significant increase of 27%, 14%, and 16%, respectively, in soil CO₂ emissions throughout the growing season. While there was no significant difference among the three crop residues, seasonal CO₂ emissions were 10% and 9% lower in the soybean and maize-soybean mixed residue treatments, respectively, compared to maize residue. Nevertheless, maize residue exhibited lower soil CO₂ emissions in the non-fertilized treatment but consistently emitted higher CO₂ in the fertilized treatments (Figure 5), indicating the significant role of N in soil CO₂ emissions from maize residue-amended soils.

In addition, the application of N fertilizers showed no significant effect on soil CO₂ emissions (Table 1). However, among fertilized treatments, manure resulted in the highest cumulative CO₂ emissions (Table 2), while maize residue under manure treatment producing the highest emissions of 8428 kg ha⁻¹ CO₂ (Figure 5). In our field study, we observed peak CO₂ fluxes following irrigation events when soil moisture was high. Consequently, a strong positive correlation was found between soil moisture and CO₂ fluxes (r = 0.31; p < 0.05) (Figure 6). Moreover, during the winter season, when the soil temperature was low, daily CO₂ fluxes were minimal, while emissions were high during autumn and spring seasons, resulting in a highly significant positive correlation between soil temperature and CO₂ fluxes throughout the study period (r = 0.79; p < 0.01). CO₂ fluxes were also strongly correlated with N₂O fluxes (r = 0.51; p < 0.01), soil NH₄⁺ (r = 0.48; p < 0.01), and DOC concentration (r = 0.30; p < 0.05).

3.6. Soil CH₄ Emissions

In our study, soils from all treatments acted as a net sink of CH₄ fluxes (Figure 5). However, both positive and negative CH₄ fluxes were recorded in daily measurements. CH₄ fluxes in crop residue amendments ranged from −37.6 to 11.3 µg m⁻² h⁻¹, while in N treatments, they ranged from −43.3 to 17 µg m⁻² h⁻¹ throughout the growing season (Figure 8). However, crop residues, N treatments, and their interactions had no significant effect on cumulative CH₄ uptake (Table 1). Nevertheless, among the N treatments, higher CH₄ uptake was observed in manure amended soils (Table 2). Manure and manure + urea amended soils’ uptake 22% and 21% more CH₄, respectively, compared to urea treatment. Among the residue amendments, soils treated with maize-soybean mixed residue showed the highest CH₄ uptake, which was 10% and 16% higher than maize and no-residue treatment, respectively.

In our study, during the winter when the soil temperature was very low, CH₄ fluxes were negative, indicating soil uptake of CH₄ from the atmosphere. However, after winter, CH₄ uptake by the soil gradually decreased with the increasing temperature, and occasionally, fluxes became positive until harvesting (Figure 8). Consequently, a significant positive correlation was found between CH₄ fluxes and soil temperature during the study period (r = 0.50; p < 0.01) (Figure 6). CH₄ fluxes also had a strong positive correlation with soil moisture content (r = 0.37; p < 0.01), soil pH (r = 0.38; p < 0.01), soil NH₄⁺ (r = 0.33; p < 0.01), and DOC concentrations (r = 0.25; p < 0.05) in the soil.

3.7. Global Warming Potential and Yield-Scaled GHG Emissions

In our results, crop residue incorporation into the soil significantly increased the global warming potential (GWP) (Table 2). Among the three crop residues, maize residue amendment showed the highest warming potential, being 25%, 9%, and 8% higher than no residue, mixed residue, and soybean residue, respectively. While all crop residue amendments led to significantly higher GWP, soybean residue and maize-soybean mixed residue exhibited significantly lower yield-scale emissions (GHGI) due to their higher grain yield. In contrast, maize residue incorporation significantly increased GHGI compared to other residue treatments, showing a 27%, 34%, and 35% higher value than no residue, mixed residue, and soybean residue, respectively.

Furthermore, in our study, there was no significant difference among the N treatments in terms of GWP. However, due to the higher grain yield, all fertilized treatments significantly reduced GHGI compared to the non-fertilized control. Among the fertilized treatments, both urea and manure + urea treatments showed similar GHGI values and were significantly lower than the sole manure treatment.

3.8. Crop Residue and N Fertilizer Effect on Soil Physicochemical Properties

In our study, crop residue addition to the soil significantly influenced soil parameters, which strongly regulated soil GHG emissions. Both NH₄⁺ and NO₃⁻ concentrations were significantly higher (p < 0.05) in soil amended with soybean residue throughout the growing season (Figure 9a,b). On average, soybean residue amendment increased soil NH₄⁺ concentration by 29%, 16%, and 13% compared to no-residue, maize residue, and mixed residue amendment, respectively. As expected, soils with maize and no-residue amendments showed significantly lower NH₄⁺ content. The soil NH₄⁺ content in the maize-soybean mixed residue treatment was intermediate and did not significantly differ from other residue amendments. Surprisingly, the average soil NO₃⁻ concentration in the mixed residue treatment (7.83 mg kg⁻¹) was the lowest among all treatments, significantly lower than the soybean residue amendment (10.26 mg kg⁻¹) (Figure 9b). Although the soil NO₃⁻ content in the soybean and maize residue amendments did not differ statistically, the average NO₃⁻ content was 27% higher in the soybean residue-amended soil compared to the maize residue amendment.

Moreover, in our study, the application of N fertilizers also increased soil NH₄⁺ concentration, but the difference was not significant compared to the non-fertilized control (Figure 10a). However, all fertilized treatments significantly enhanced soil NO₃⁻ concentration (Figure 10b). Average NO₃⁻ concentration was highest in the urea fertilized soil, which was 38% and 48% higher than manure and manure + urea treated soil, respectively.

Soil DOC concentration was also influenced by both crop residue and N fertilizer treatments. Throughout the growing season, soil DOC concentration was significantly higher in maize residue-amended soil (Figure 9d). While all N fertilizers increased soil DOC concentration, the increment was notably higher in the manure-treated soil (Figure 10d). Soil temperature, moisture, C/N ratio, and pH content were not significantly changed by the crop residue incorporation. However, all N fertilizers significantly reduced soil pH content (Figure 10h), while the soil C/N ratio was significantly lower in the manure-treated soil throughout the study period (Figure 10c).

4. Discussions

4.1. Crop Residue and Fertilizer Effect on Soil N Dynamics and N₂O Emissions

In our study, crop residue incorporation into the soil notably increased soil N₂O emissions. However, the quality of the residue determined the magnitude of N₂O emissions and soil N dynamics. Particularly, the incorporation of soybean residue significantly increased the soil mineral N concentration, with soil NH₄⁺ and NO₃⁻ concentrations being 16% and 27% higher, respectively, compared to maize residue-amended soil (Figure 9a,b). Such an increment in NH₄⁺ and NO₃⁻ concentrations in the soil could be attributed to the low C/N ratio and rapid mineralization of biologically fixed N in the soybean residue and root nodules [39]. This observation of high mineral N in the soybean residue-amended soil is consistent with previous findings indicating that incorporating easily degradable residues with a low C/N ratio increases soil mineral N content more than applying residues with a high C/N ratio [11,40,41]. Consequently, the availability of NH₄⁺ and NO₃⁻ concentrations enhanced nitrification and denitrification in the soil, leading to higher N₂O emissions from soybean residue-amended soil.

Conversely, the incorporation of maize residue and maize-soybean mixed residue reduced both NH₄⁺ and NO₃⁻ concentrations in the soil. This reduction could be attributed to the slower decomposition and slower N mineralization processes in the soil due to the high C/N ratio and high lignin content of the maize residue [42]. Another possible reason is that, due to its low N content, maize residue cannot meet the N requirement of microbial growth. As a result, heterotrophic microorganisms assimilate indigenous soil N to decompose organic C in maize residue, causing net N immobilization [4,41,43,44]. Consequently, both maize residue and maize-soybean mixed residue-amended soil showed lower NH₄⁺ and NO₃⁻ concentrations (Figure 9a,b) and subsequently lower soil N₂O emissions. Previous studies have also reported a significant negative correlation between crop residue C/N ratio and soil N₂O emissions [14,45,46]. A meta-analysis by Chen et al. [4] indicated that incorporating crop residue with a C/N ratio < 45 significantly increases soil N₂O emissions. When crop residue has a lower C/N ratio, it decomposes faster after incorporation into the soil, providing sufficient N to support the growth and proliferation of the soil microbial community, leading to higher N availability for nitrification and denitrification [46,47]. Therefore, soil amended with soybean residue had 27% higher N₂O emissions than soil amended with maize residue. However, in our study, despite the presence of soybean residue, N₂O emissions were lower in the mixed residue-amended soil. This could be because large volume of maize residue in the mixture resulted in an overall high C/N ratio.

In our experiment, seasonal temperature variation also notably affected soil N dynamics. A few weeks after the residue incorporation, soil NH₄⁺ decreased but NO₃⁻ content increased (Figure 3). This change may have occurred due to the nitrification of NH₄⁺ and reduced decomposition rate, as well as the slower mineralization of organic N during winter [48]. However, during spring, with the increasing soil temperature, soil NH₄⁺ concentration sharply increased until the maturity stage, indicating the temperature effect on residue decomposition and organic N mineralization. In our experiment, both soil NH₄⁺ and NO₃⁻ concentrations were significantly and positively correlated with soil temperature (r = 0.22; p < 0.01). Many incubation studies have shown that soil N mineralization exponentially increases with the increasing soil temperature [48,49,50]. This increase in N mineralization can be explained by the rapid consumption of readily mineralizable organic matter [51] or as a result of greater depolymerization of organic matter due to the more intense activity of exocellular enzymes at elevated temperatures [52]. During the maturity stage of winter wheat, soil NH₄⁺ concentration decreased again with an increase in NO₃⁻ content, possibly due to the low N requirement of the plants during maturity stage but the continuous nitrification as well as recycling of N with microbial death [53], leading to an increase in soil NO₃⁻ content at the maturity stage.

Elevated temperatures also enhance soil nitrification and denitrification rates, subsequently increasing soil N₂O emissions [54,55]. In our study, N₂O emissions also demonstrated a significant positive correlation with soil temperature (r = 0.32; p < 0.01) (Figure 6). Under elevated temperature, soil nitrification rates increase due to an increase in the abundance of ammonia-oxidizing archaea (AOA) or ammonia-oxidizing bacteria (AOB) amoA genes [56]. Furthermore, elevated soil temperature increases the abundance of the denitrifying gene nirK, thereby influencing soil denitrification [55,56]. In addition, an increase in soil temperature increases soil microbial respiration, leading to a depletion of oxygen concentrations in the soil and creating anaerobic microsites, which favor denitrification and N₂O emissions [12,57] Therefore, a significant positive correlation (r = 0.51; p < 0.01) was observed between soil CO₂ and N₂O emissions in this study. Similar correlations have also been recorded in some previous studies when plant materials were incorporated into the soil [11,55,58]. This is because, in addition to N, crop residue incorporation increases available organic C, which stimulates soil heterotrophic respiration and the oxidation of labile organic C fractions to CO₂, lowering oxygen pressures in soils and creating anaerobic conditions favorable for denitrifiers [59,60].

Although N fertilizer application in agricultural soils boosts grain yield, it is also known to cause a notable rise in soil N₂O emissions [61]. In our study, all N fertilizers significantly increased soil N₂O emissions, with the highest N₂O emissions observed in urea-fertilized soil (Table 2). In a field experiment, Mairura et al. [28] showed that substituting urea with manure could reduce soil N₂O emissions. Similarly, in our study, the manure + urea treatment had lower N₂O emissions than the sole urea amendment. However, both the urea and manure + urea treatments had significantly higher N₂O emissions than the sole manure treatment. This is because, compared to manure, urea has a high denitrification potential [27]. Moreover, urea application improves N availability to the microbial community, speeds up the decomposition of plant materials, enhances the NH₄⁺ and NO₃⁻ concentrations in the soil, and subsequently increases N₂O emissions [62]. Reduced N₂O emissions from manure-treated soil are attributed to the gradual release of plant-accessible N throughout the growing season from the manure [63]. Consequently, once the plants’ N needs are met, there remains a lesser amount of mineral N accessible in the soil. Therefore, throughout the growing season, both manure and manure + urea amended soil had significantly lower NO₃⁻ content than urea amended soil (Figure 10b), resulting in lower N₂O emissions in both manure-treated soils.

4.2. Crop Residue and N Fertilizer Effect on Soil CO₂ Emissions

Our results showed that crop residue incorporation significantly increased soil CO₂ emissions compared to no residue amendment. Similar results were also observed in previous studies on CO₂ emissions from residue-amended soils [11,13,53]. This is because crop residue serves as a significant source of external carbon (C) input, which alters soil microbial activity and subsequently affects native SOC mineralization (priming effect) [64,65]. Crop residue often enhances the positive priming effect through processes like the “co-metabolism” or “microbial nutrient mining” of native soil organic matter, consequently increasing soil CO₂ emissions [66]. Additionally, the incorporation of crop residue triggers microbial activity in the soil, accelerating internal microbial metabolism and increasing respiratory activity immediately upon residue incorporation [58].

However, the magnitude of soil CO₂ emissions from maize residue-amended soil varied under different N treatments. Specifically, soil CO₂ emissions were relatively low when maize residue was incorporated into non-fertilized soil but increased dramatically compared to other residues when incorporated into N-fertilized soils (Figure 5). This suggests that lower microbial activity occurs in maize residue-amended soil compared to soybean and maize-soybean mixed residue when there is a shortage of inorganic N in the soil, indicating that soil available N is a limiting factor for microbial biomass, activity, and residue decomposition [43,67,68]. It has been reported that a shortage of N, rather than a shortage of C, in residue-amended soils would limit microbial activity during the decomposition process [69,70]. Moreover, when plenty of fresh organic materials with a high C/N ratio are incorporated into N-limited soil, microbial N demand cannot be met from either source, resulting in lower microbial activity and CO₂ emissions [71]. Therefore, during the basal fertilization period when N uptake by plants is negligible, the applied N fertilizer provides adequate nutrients to microorganisms for residue decomposition, resulting in high CO₂ fluxes in residue-amended soils and N-fertilized treatments during the first two weeks (Figure 7). Similar results were also observed during the top-dressing period.

Overall, the C priming effect is generally higher in low C/N residue-amended soil than in residue with a high C/N ratio in the early stage of incorporation [72]. However, Brenzinger et al. [73] noted that the magnitude of soil CO₂ emissions mostly depends on the amount of organic residue added. Chen et al. [13] found that soil CO₂ emissions were significantly and positively correlated with the amount of residue added to the soil. Similarly, Liang et al. [74] reported that the soil priming effect positively correlated with the amount of residue incorporation. In our field study, the quantity of crop residue played a crucial role in terms of soil CO₂ emissions. Specifically, the amounts of incorporated maize residue (9.1 t ha⁻¹) and maize-soybean mixed residue (7.1 t ha⁻¹) were much higher than that of soybean residue (3.04 t ha⁻¹). With the increase in crop residue, enzyme activity and labile organic carbon fraction (e.g., DOC) content increased in the soil [13]. In our study, the DOC content in maize residue-amended soil was significantly higher than in other residue amendments (Figure 9d) and had a strong positive correlation (r = 0.30; p < 0.05) with soil CO₂ emissions (Figure 6), resulting in 11% more soil CO₂ emissions from maize residue than from soybean residue. Similarly, significantly higher DOC content was found in manure-amended soils in our study, leading to high CO₂ emissions in this treatment. Manure application likely provides potentially more soluble carbon for microbial activity than urea fertilizer, thus leading to greater soil CO₂ emissions [75].

Temperature variation exerted a significant effect on soil CO₂ emissions in our field study. During autumn, as temperature decreased, CO₂ emissions gradually decreased, remained low during winter, and increased again during spring with rising soil temperature, resulting in a significant positive correlation (r = 0.79; p < 0.01) between soil temperature and soil CO₂ emissions. This temperature effect on soil CO₂ emissions aligns with findings from previous studies [54,65,76]. Generally, soil temperature strongly regulates microbial functioning in the soil. Elevated soil temperatures accelerate soil microbial respiration and DOC concentration but lower microbial biomass C [55]. The positive correlation between soil temperature and DOC concentration (r = 0.40; p < 0.01) (Figure 6) in our study may partly explain the temperature effect on higher soil CO₂ emissions. It has been reported that at a temperature of 25 °C, soil microbial activity is high, with a greater amount of mineralizable (labile) C in the soil [76,77]. Irina et al. [78] also reported that microbial decomposition of residue and soil CO₂ emissions significantly increased with increasing temperature up to 25 °C, supporting our results of increased CO₂ emissions during spring.

Throughout our study period, soil CO₂ fluxes were high after irrigation or sudden rainfall events. Thus, soil CO₂ fluxes showed a strong positive correlation with soil moisture content (r = 0.31; p < 0.05). Soil moisture content also had a positive association with soil DOC content (r = 0.27; p < 0.05). There is a trend of increasing DOC content following re-wetting after dry spells [79]. In upland agriculture, soils often experience periodic dry periods and accumulate microbial products due to reduced decomposition rates. When irrigation is applied or rainfall occurs, soil DOC concentration rises, subsequently consumed by soil microbes as substrates, leading to increased soil CO₂ emissions [80,81]. Moreover, when soil moisture is limited, microbial activity is very low, or microorganisms die due to water scarcity [82]. Conversely, when soils have sufficient moisture, residue decomposition and SOC mineralization rates increase with increased microbial activity, resulting in high CO₂ emissions.

4.3. Crop Residue and N Fertilizer Effect on CH₄ Emissions

In our study, soils across all treatments acted as a net CH₄ sink. Perhaps this was because our experiment was conducted in an upland field that was mostly aerobic, and the crop residue incorporation further increased soil aeration [83,84]. Despite this, neither crop residue amendments nor N fertilizers significantly impacted the cumulative soil CH₄ uptake. However, CH₄ fluxes varied among sampling events throughout the growing seasons, largely influenced by environmental factors. Factors such as soil temperature, moisture content, NH₄⁺, and DOC concentrations likely contributed to the observed variation in net CH₄ fluxes (Figure 6).

Compared to the basal fertilizer application period, the highest CH₄ emissions occurred during the top dressing of fertilizer, coinciding with the irrigation events. Positive CH₄ fluxes or lower CH₄ uptake were also observed during other irrigation or rainfall events or periods of high soil moisture content, indicating that soil moisture was the instrumental factor for CH₄ fluxes. Additionally, during spring, two irrigation events of 65 mm and 70 mm were applied, with comparatively more rainfall occurring during this period. This likely contributed to lower CH₄ consumption or even CH₄ emissions during the second half of the growing season compared to the first half. Our correlation analysis also revealed a significant positive correlation between CH₄ emissions and soil moisture content (r = 0.37; p < 0.01). Elevated soil moisture content during irrigation and rainfall restricts oxygen diffusion into the soil, creating an anaerobic environment that inhibits methanotrophic activities (the oxidation of CH₄) while promoting methanogenic activities responsible for CH₄ production [85].

In our study, manure application enhanced CH₄ uptake compared to urea fertilizer. A similar trend was also observed in previous studies, where manure amendment significantly increased CH₄ uptake in upland soils [86,87]. Manure application generally increases soil dissolved carbon, which could facilitate the activity of methane-oxidizing bacteria and thus CH₄ oxidation [88,89].

Furthermore, soil temperature exerted a significant negative effect (p < 0.01) on soil CH₄ consumption in our study. With the increasing temperature during spring, soil CH₄ consumption dramatically decreased, and even occasionally, CH₄ emissions were positive when coinciding with elevated soil moisture. The optimum temperature for methanogenesis is typically between 30 and 40 °C, while methanotrophy is less sensitive to temperature than methanogenesis [90]. Low soil temperatures reduce CH₄ production by decreasing soil methanogenic bacterial activities, whereas methanogenic activities increase several fold with increasing temperatures from 5 to 35 °C [91,92]. A recent study by Fan et al. [93] also showed that the rates of CH₄ oxidation and methanogenesis increased exponentially with temperature, but the CH₄ oxidation rate was significantly lower than methanogenesis when temperature reached 35 °C, resulting in more CH₄ production compared to CH₄ oxidation.

4.4. Yield-Scaled Emissions and Potentials of Maize-Soybean Mixed Residue

In our study, we observed that soybean residue amendment significantly increased soil N₂O emissions, while maize residue had the highest soil CO₂ emissions. However, the higher N₂O emissions from soybean residue and the higher CO₂ emissions from maize residue could be mitigated using maize-soybean mixed residue. Our results indicated that maize-soybean mixed residue could reduce soil N₂O emissions by 23% compared to soybean residue and soil CO₂ emissions by 10% compared to maize residue. Additionally, mixed residue amendment also contributed to a significantly higher grain yield compared to no residue addition.

Previous studies have reported numerous advantages of cereal-legume intercropping systems, mainly higher grain yield [94,95,96] and lower GHG emissions due to the low N requirement in the system [97,98,99], suggesting intercropping as a promising cropping system for sustainable agroecosystems. However, concerns have been raised regarding higher GHG emissions, particularly N₂O emissions, after residue incorporation into the soil due to the presence of legume residue in the mixture, as legume residue significantly increases soil N₂O emissions [100,101]. In our study, however, maize-soybean mixed residue produced significantly lower N₂O emissions compared to soybean residue and exhibited the lowest GWP among all the residues. Most importantly, with the increasing yield, mixed residue significantly reduced yield-scaled emissions in the winter wheat (Table 2). While previous studies have primarily focused on area-scaled GHG emissions, considering the GHG efficiency of crop production suggests that yield-scaled GHG emissions from agro-ecosystems are likely to provide a more accurate measurement of GHG emissions [25]. In our study, maize-soybean mixed residue significantly reduced yield-scaled emissions by 25% compared to maize residue amendment. This reduction of GHGI from cereal-legume mixed residue could encourage policymakers to promote diversified cropping systems like intercropping over current high-fertilized monocropping systems.

5. Conclusions

Results from our study showed that crop residue addition significantly influenced soil GHG emissions, particularly soil N₂O and CO₂ emissions, depending on the type and quantity of residue biomass added to the soil. Comparing three different crop residues with varying C/N ratios, we found that as soil inorganic N (NH₄⁺ and NO₃⁻) content increased, soybean residue led to significantly higher N₂O emissions, while maize residue was responsible for the highest CO₂ emissions. In contrast, maize-soybean mixed residue, derived from an intercropping system, exhibited the lowest N₂O emissions among the three residues and notably lower CO₂ emissions compared to maize residue, resulting in the lowest GWP. Additionally, mixed residue amendment contributed to higher grain yield and significantly reduced GHGI by 25% compared to maize residue amendment. Overall, our findings contribute to a comprehensive understanding of how different residue additions, from monocrops and cereal-legume intercrops, influence soil N dynamics and GHG emissions, providing valuable insights into effective agroecosystem management for long-term food security and soil sustainability while mitigating GHG emissions. Moreover, using manure instead of synthetic N fertilizer also reduced soil N₂O emissions. However, the strategic application of N fertilizer, particularly urea as a top dressing during the tillering stage, remains crucial for achieving higher wheat grain yields. In our study, the manure + urea treatment exhibited a lower GWP but produced the highest grain yield and subsequently the lowest GHGI among the treatments. Therefore, partially substituting urea with manure as a basal application could effectively reduce soil GHG emissions while maintaining productivity.