Mitigating Carbon and Nitrogen Footprints While Enhancing Ecosystem Economic Benefits via Strategic Application of Slow-Release Fertilizer and Mulching

Abstract

Dryland farming on the Loess Plateau faces significant challenges due to water scarcity and low nitrogen use efficiency. Although conventional urea sustains crop yields, it is associated with elevated greenhouse gas emissions and nitrogen losses. Despite growing interest in both slow-release fertilizers and plastic mulching, their environmental footprints remain insufficiently evaluated. This study, therefore, aimed to identify a management strategy that maximizes productivity while minimizing environmental impacts. Using a life-cycle assessment (LCA) framework, we compared four cultivation strategies, flat cultivation with urea (NU), flat cultivation with slow-release fertilizer (NS), mulching with urea (PU), and mulching with slow-release fertilizer (PS), each at nitrogen rates of 125, 225, and 325 kg ha⁻¹. The results demonstrated that PS reduced the carbon footprint per unit of net ecosystem economic benefits (NEEB) by 3.74–27.86% and the nitrogen footprint per unit of NEEB by 10.48–47.41%. At 225 kg ha⁻¹, PS increased grain yield and NEEB by 7.40% and 9.87%, respectively, compared to 125 kg ha⁻¹. Compared to 325 kg ha⁻¹, the 225 kg ha⁻¹ rate improved energy use efficiency by 19.81% while reducing carbon emissions, carbon, and nitrogen footprint per unit of NEEB by 10.29%, 14.36%, and 24.47%, respectively. In conclusion, mulching combined with slow-release fertilizer at 225 kg ha⁻¹ represents a balanced and regionally appropriate strategy, achieving strong agronomic performance alongside reduced environmental costs and improved economic returns.

1. Introduction

The rapid increase in greenhouse gas (GHG) emissions is accelerating global warming, and agriculture, which accounts for approximately one-quarter of global emissions, represents a significant contributor to atmospheric carbon emissions [1,2]. Beyond GHG emissions, reactive nitrogen (RN) losses such as N₂O, NH₃, NO₃⁻, and NH₄⁺ have emerged as critical indicators of the environmental impacts of agricultural activities [3]. To mitigate these effects, it is crucial to assess and manage the ecological impacts of agricultural inputs using environmental indicators such as carbon and nitrogen footprints (GHG emissions and RN losses per unit of grain yield). However, these environmental impacts are shaped not only by management inputs but also by physical factors and biogeochemical processes regulating carbon and nitrogen transformations. Such interactions generate pronounced spatial and temporal variability, often manifested as emission hotspots and episodic pulses [4,5,6]. A comprehensive assessment framework that accounts for these complexities is therefore essential to reduce environmental costs while maintaining food security [7,8].

Spring maize is the dominant staple crop on the Loess Plateau and plays a central role in regional food security and agricultural economic development [9,10]. Its cultivated area has expanded steadily and now accounts for more than 30% of the total grain cropping area, surpassing traditional crops such as winter wheat. This expansion is largely attributable to its comparatively high water use efficiency and yield potential under semi-arid conditions. Nevertheless, agricultural productivity in the region remains constrained by limited and uneven precipitation (300–550 mm annually, with 60–70% concentrated between July and September), restricted groundwater availability, and inherently poor soil fertility [11].

To mitigate water limitations, plastic film mulching has been widely adopted across the Loess Plateau. By increasing soil temperature and conserving moisture, mulching enhances crop establishment and yield stability in dryland systems [12,13]. However, prolonged and intensive plastic use may negatively affect soil quality by altering organic matter dynamics, disrupting decomposition processes, and potentially increasing soil GHG emissions [14,15,16]. Balancing the agronomic benefits of mulching with its long-term environmental consequences has therefore emerged as a research priority [17,18,19].

Alongside water management, optimizing nitrogen input is equally essential for improving both productivity and environmental performance. In film mulching systems, fertilizer type and application rate strongly influence nitrogen use efficiency and environmental losses [20,21,22]. Slow-release fertilizers, designed to synchronize nutrient release with crop demand, have been shown to reduce nitrogen leaching and ammonia volatilization, while sustaining high yields [23]. By improving temporal matching between nitrogen supply and plant uptake, SRFs can potentially reduce total nitrogen inputs compared to conventional urea. In contrast, excessive nitrogen application exacerbates reactive nitrogen losses and GHG emissions, while insufficient inputs constrain productivity [24,25,26]. Determining an appropriate nitrogen rate is therefore fundamental to achieving agronomic and environmental balance.

Although numerous studies have demonstrated that mulching improves spring maize yields in semi-arid regions of the Loess Plateau [12,27,28,29], relatively few have systematically examined how fertilizer type and nitrogen rate interact under mulched, rainfed conditions to influence yield, energy efficiency, carbon footprint, nitrogen footprint, and net ecosystem economic benefit (NEEB). In particular, quantitative evidence identifying a nitrogen rate that simultaneously maximizes economic returns and minimizes environmental costs in this system remains limited. Therefore, this study aims to: (1) evaluate the effects of different nitrogen fertilizer management strategies on yield, environmental footprints, and NEEB in film-mulched systems, (2) analyze energy inputs, outputs, and energy use efficiency under contrasting practices, and (3) propose a regionally appropriate management strategy that balances food production, economic performance, and environmental sustainability in the dryland agroecosystems of the Loess Plateau.

2. Materials and Methods

2.1. Experimental Site

This experiment was conducted from 2019 to 2022 at the Changwu Ecological Agricultural Experimental Station (35°14′ N, 107°41′ E; 1200 m a.s.l.), located in Changwu County, Xianyang City, Shaanxi Province, China, within the Weibei Dry Plateau region of the Loess Plateau (Figure 1). The site has a warm-temperate, semi-humid continental monsoon climate. During the experimental period, the mean annual temperature was 10.74 °C and total annual rainfall precipitation was 598.93 mm, of which 496.33 mm occurred during the maize growing season (Figure 2). The frost-free period averaged 171 days.

Figure 1. Geographic location of the Changwu Ecological Agricultural Experimental Station (35°14′ N, 107°41′ E; 1200 m a.s.l), situated in Changwu County, Xianyang City, Shaanxi Province, China.

Figure 2. Total rainfall (mm), maximum temperature (°C), mean temperature (°C), and minimum temperature (°C) at the experimental site from 2019 to 2022.

Before the initiation of the experiment (2019), soil properties in the 0–20 cm layer included 7.60 g kg⁻¹ soil organic carbon (SOC), 0.85 g kg⁻¹ total nitrogen, a bulk density of 1.05 g cm⁻³, and a pH of 8.44. The soil is classified as Malan loess (medium loam texture), characterized by good permeability and water retention, and is widely cultivated across the Loess Plateau.

2.2. Experimental Design

The experiment employed a three-factor randomized complete block design with four replicates. The factors included mulching practice (two levels: no mulching and plastic film mulching), nitrogen fertilizer type (two levels: conventional urea and slow-release fertilizer), and nitrogen fertilizer rate (three levels: 125, 225, and 325 kg N ha⁻¹), resulting in 12 treatments. Individual plot size was 4.3 m × 3.5 m. The treatments were flat cultivation with urea (NU), flat cultivation with slow-release fertilizer (NS), mulching with urea (PU), and mulching with slow-release fertilizer (PS).

The slow-release fertilizer was a dual-coated compound fertilizer (Shandong Linyi Woli Compound Fertilizer Co., Ltd.) consisting of an inner biological coating and an outer polymer coating designed to synchronize nutrient release with the crop demand. Its nutrient composition was 28% N, 6% P₂O₅, and 6% K₂O (w/w). Conventional urea contained 46% N. According to product specifications, cumulative nutrient release at 25 °C was ≤15% within 24 h, ≤75% within 28 days, and ≥75% over the designated release period [30,31]. To maintain a consistent N: P₂O₅: K₂O ratio across treatments, supplemental phosphorus and potassium were added to urea treatments to match the slow-release fertilizer nutrient proportions. Fertilizer inputs are detailed in Table S1. All fertilizers were applied as basal dressing before sowing, incorporated into the 0–20 cm soil layer through rotary tillage. The experimental field followed a continuous spring maize cropping system.

In late April each year, maize (cv. ‘Xianyu 335’) was sown at a density of 62,500 plants ha⁻¹. In mulching treatments, the plots were covered with transparent polyethylene film (100 cm width, 0.1 mm thickness, non-perforated) before sowing. The film was laid flat, secured at the edges, and maize was manually planted through holes punched in the film. Maize was harvested in September, and the residual film was removed in October. The system was rainfed throughout the study. Herbicides and pesticides were applied only in the unmulched treatments as required.

2.3. Measurements and Calculations

2.3.1. Grain Yield

In each plot, 15 representative maize ears were randomly sampled to determine yield components (rows per ear, grains per row, and 100-grain weight), consistent with established regional protocols [10,13]. The 100-grain weight was measured after oven-drying the grains in a forced-air oven at 75 °C to a constant weight. The final grain yield was adjusted to 14% standard moisture content [9].

2.3.2. Life-Cycle Assessment

Environmental impacts were quantified using a cradle-to-farm-gate life-cycle assessment (LCA) approach. The system boundary included upstream processes (raw material extraction, fertilizer and film production, transportation) and on-farm operations (machinery use and field emissions). Indirect emissions were calculated by multiplying input quantities by localized emission factors. Grain yield was the functional unit, and no allocation was required.

Carbon (CF_Yield) and nitrogen (NF_Yield) footprints were calculated by normalizing total greenhouse gas (GHG) emissions and reactive nitrogen (RN) losses to grain yield. Emission factors were prioritized from localized studies in the Loess Plateau and national guidelines (including IPCC standards) to ensure regional relevance (Table S2).

Carbon Footprint

Greenhouse gas emissions included those from fertilizer production and transportation, mulch, diesel, direct and indirect emissions from nitrogen fertilizer, and decomposition of maize roots and stubble [32,33,34]. Key metrics included net GHG emissions (GHG_Net, kg CO₂-eq.kg ha⁻¹), total GHG emissions (GHG_Total, kg CO₂-eq.kg ha⁻¹), emissions from nitrogen fertilizer application (GHG_N, kg CO₂-eq.kg ha⁻¹), emissions from maize roots (GHG_R, kg CO₂-eq.kg ha⁻¹), changes in emissions from organic carbon storage (GHG_SOC, kg CO₂-eq.kg ha⁻¹), and carbon footprint per unit of yield (CF_Yield, kg CO₂-eq.kg⁻¹), calculated as follows:

$${\mathrm{GHG}}_{\mathrm{Net}}={\mathrm{GHG}}_{\mathrm{Total}}-{\mathrm{GHG}}_{\mathrm{SOC}}$$

(1)

$${\mathrm{GHG}}_{\mathrm{Total}}=\mathsf{\Sigma }{\mathrm{I}}_{\mathrm{n}}\times {\mathrm{EF}}_{\mathrm{n}}+{\mathrm{GHG}}_{\mathrm{N}}+{\mathrm{GHG}}_{\mathrm{R}}$$

(2)

$${\mathrm{GHG}}_{\mathrm{N}}={\mathrm{N}}_{\mathrm{F}}\times \left({\mathrm{EF}}_{\mathrm{FN}}+{{\mathrm{F}\mathrm{r}\mathrm{a}\mathrm{c}}_{\mathrm{v}\mathrm{o}\mathrm{l}}\times \mathrm{EF}}_{\mathrm{vol}}+{{\mathrm{F}\mathrm{r}\mathrm{a}\mathrm{c}}_{\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{h}}\times \mathrm{EF}}_{\mathrm{leach}}\right)\times 44/28\times 298$$

(3)

$${\mathrm{GHG}}_{\mathrm{R}}={\mathrm{N}}_{\mathrm{R}}\times \left({\mathrm{EF}}_{\mathrm{RN}}+{{\mathrm{F}\mathrm{r}\mathrm{a}\mathrm{c}}_{\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{h}}\times \mathrm{EF}}_{\mathrm{leach}}\right)\times 44/28\times 298$$

(4)

$${\mathrm{GHG}}_{\mathrm{SOC}}=\left({\mathrm{SOC}}_{2022}\times \mathrm{h}\times \mathsf{\rho }\times {10}^4-{\mathrm{SOC}}_{2019}\times \mathrm{h}\times \mathsf{\rho }\times {10}^4\right)/4\times 44/12$$

(5)

$${\mathrm{CF}}_{\mathrm{Yield}}={\mathrm{G}\mathrm{H}\mathrm{G}}_{\mathrm{N}\mathrm{e}\mathrm{t}}/\mathrm{Yield}$$

(6)

where N_F and N_R represent the quantity of N applied as inorganic fertilizer (kg N ha⁻¹) and maize residue nitrogen (kg N ha⁻¹). I_n and EF_n represent agricultural inputs and emission factors during maize production (Tables S1 and S2). The emission factors and coefficients were prioritized from localized studies in the Loess Plateau or national standards (IPCC) to ensure regional relevance. Direct N₂O emissions from nitrogen fertilizer were calculated by multiplying nitrogen application rates by specific emission factors (EF_FN). The base emission factor for urea (EF_FN-Urea) was rectified using total precipitation during the maize growing period (P) as 0.0186 × P kg N₂O-N kg⁻¹ [35]. For treatments using slow-release fertilizer, a specialized emission factor (EF_FN-SRF) was applied, calculated as 42.51% of the urea-based factor (0.4251 × EF_FN-UREA, kg N₂O-N kg⁻¹) [36]. Frac_vol represents the fraction of fertilizer nitrogen volatilized as NH₃-N and NO_x-N, while Frac_leach represents the fraction of fertilizer and maize residue nitrogen for NO₃⁻ leaching; EF_vol and EF_leach represent N₂O emission factor for volatilized NH₃-N and NO_x-N and nitrate leaching; the above values related to volatilization and leaching, as well as the corresponding references, were all shown in Table S2. EF_RN represents the emission factor for maize residue nitrogen (0.0059 P, kg N kg⁻¹) [37], SOC₂₀₂₂ and SOC₂₀₁₉ represent soil organic carbon content (0–40 cm) in 2022 and 2019, respectively, which were determined using the potassium dichromate (K₂Cr₂O₇) oxidation method. In this study, 44/28 is the conversion coefficient (N₂O-N to N₂O), 44/12 is the carbon conversion coefficient (CO₂ to organic carbon), 298 is the global warming potential for N₂O for 100 years [38], h is soil depth (m), and ρ is soil bulk density (g cm⁻³).

Nitrogen Footprint

Nitrogen loss components included NH₃ volatilization, direct emission of N₂O, and NO₃⁻ and NH₄⁺ leaching. Metrics were calculated using established formulas [39]. Key metrics included RN losses (kg N-eq ha⁻¹), RN losses due to volatilization of nitrogen fertilizer (N-NH₃, kg N-eq ha⁻¹), RN losses due to N₂O emissions (N-N₂O, kg N-eq ha⁻¹), RN losses due to NO₃⁻-N leaching (N-NO₃⁻, kg N-eq ha⁻¹), RN losses due to NH₄⁺-N leaching (N-NH₄⁺, kg N-eq ha⁻¹), and the nitrogen footprint per unit of yield (NF_Yield, kg N Mg⁻¹) can be calculated using the following formulas:

$$\mathrm{RN\; losses}=\mathsf{\Sigma }{\mathrm{I}}_{\mathrm{n}}\times {\mathrm{NF}}_{\mathrm{n}}+\mathrm{N}\text{-}{\mathrm{NH}}_3+\mathrm{N}\text{-}{\mathrm{N}}_2\mathrm{O}+\mathrm{N}\text{-}{\mathrm{NO}}_{3^{-}}+\mathrm{N}\text{-}{\mathrm{NH}}_{4^{+}}$$

(7)

$$\mathrm{N}\text{-}{\mathrm{NH}}_3={\mathrm{N}}_{\mathrm{F}}\times {\mathrm{Frac}}_{{\mathrm{volNH}}_3}\times {\mathrm{N}\mathrm{F}}_{\mathrm{v}\mathrm{o}\mathrm{l}}\times 17/14$$

(8)

$$\ln {\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}_{{\mathrm{N}}_2\mathrm{O}}=0.0045\times {\mathrm{N}}_{\mathrm{F}}+0.0742\times \mathrm{T}\mathrm{e}\mathrm{m}\mathrm{p}+0.0134\times \mathrm{C}\mathrm{l}\mathrm{a}\mathrm{y}-2.0092$$

(9)

$$\mathrm{N}\text{-}{\mathrm{N}}_2\mathrm{O}={\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}_{{\mathrm{N}}_2\mathrm{O}}\times 44/28\times 0.476$$

(10)

$$\mathrm{N}\text{-}{\mathrm{NO}}_{3^{-}}={\mathrm{N}}_{\mathrm{F}}\times {\mathrm{Frac}}_{\mathrm{leachNO}{3}^{-}}\times 62/14\times {\mathrm{N}\mathrm{F}}_{\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{h}}$$

(11)

$$\mathrm{N}\text{-}{\mathrm{NH}}_{4^{+}}={\mathrm{N}}_{\mathrm{F}}\times {\mathrm{Frac}}_{\mathrm{leachNH}{4}^{+}}\times 18/14\times {\mathrm{N}\mathrm{F}}_{\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{h}}$$

(12)

$${\mathrm{NF}}_{\mathrm{Yield}}=\mathrm{RN\; losses}/\mathrm{Yield}$$

(13)

where NF_n represents RN_Loss per unit of agricultural input (10⁻³ kg N unit⁻¹). N_F represents the quantity of N applied as inorganic fertilizer (kg N ha⁻¹). Total_N2O is calculated according to Yue et al. [40]. Temp means the annual average temperature (°C); Clay indicates the fraction of clay (%). Frac_volNH3, Frac_leachNO3− and Frac_leachNH4+ represent the fraction of RN_Loss volatilized as NH₃-N, RN_Loss for NO₃⁻ leaching, and RN_Loss for nitrate leaching; NF_vol and NF_leach represent reactive nitrogen loss coefficients for volatilized NH₃-N and nitrate leaching; the above values related to volatilization and leaching, as well as the corresponding references, were all shown in Table S2. 17/14, 44/28, 62/14, and 18/14 are conversion factors for N-NH₃, N-N₂O, N-NO₃⁻, and N-NH₄⁺, respectively. Slow-release fertilizers can significantly reduce nitrogen oxide emissions, nitrogen leaching, and ammonia volatilization by 23.8%, 27.1%, and 39.4%, respectively [41]. Other factors and coefficients are detailed in Table S2.

2.3.3. Energy Use Efficiency

Energy inputs comprised direct (e.g., diesel and labor) and indirect sources (e.g., fertilizers, mulch). Outputs included energy from grain and stalks. Renewable (e.g., labor, seeds) and non-renewable inputs (e.g., diesel, fertilizers) were categorized. Key metrics include net output energy (NE, 10⁻³ MJ ha⁻¹) and energy use efficiency (EUE, kg MJ⁻¹), calculated as follows:

$$\mathrm{NE}=\mathrm{Oe}-\mathrm{Ie}$$

(14)

$$\mathrm{EUE}=\mathrm{Yield}/\mathrm{Ie}$$

(15)

where Oe and Ie represent the total output and input energies of the agricultural production system, respectively. Conversion factors for input and output materials are detailed in Table S3.

2.3.4. Net Ecosystem Economic Benefits

The net ecosystem economic benefit (NEEB) was calculated on agricultural input costs, carbon emission costs, and income from food production [42]. Key metrics included NEEB (CNY ha⁻¹ year⁻¹), carbon footprint per unit of NEEB (CF_NEEB, kg CO₂-eq.CNY⁻¹ year⁻¹), and nitrogen footprint per unit of NEEB (NF_NEEB, kg N CNY⁻¹ year⁻¹), calculated as follows:

$$\mathrm{NEEB}=\mathrm{GR}-{\mathrm{AI}}_{\mathrm{cost}}-{\mathrm{GHG}}_{\mathrm{cost}}$$

(16)

$${\mathrm{CF}}_{\mathrm{NEEB}}={\displaystyle \frac{{\mathrm{GHG}}_{\mathrm{Net}}}{\mathrm{NEEB}}}$$

(17)

$${\mathrm{NF}}_{\mathrm{NEEB}}={\displaystyle \frac{\mathrm{RN\; losses}}{\mathrm{NEEB}}}$$

(18)

where GR represents the benefits obtained from food production, AI_cost represents the costs generated by fertilizers and films invested in agricultural production (see Tables S1 and S4 for related inputs and prices), and GHG_cost represents the costs generated by carbon emissions (carbon trading price multiplied by GHG emissions).

2.4. Statistical Analysis

Statistical analyses and data visualization were conducted to evaluate the effects of nitrogen management on crop productivity and environmental indicators. Analysis of variance (ANOVA) was performed with SPSS 22.0 (IBM, Chicago, IL, USA), and the least significant difference (LSD) method was employed for multiple comparisons among treatment means at a significance level of p < 0.05. Other figures, including treatment comparisons and trend analyses, were generated using Origin 2018 (Origin Lab, Northampton, MA, USA). The study site map and geographical distributions were created using ArcGIS 10.8 (ESRI, Redlands, CA, USA).

3. Results

3.1. Maize Grain Yield

The precipitation in this region from 2019 to 2022 was primarily concentrated between May and October, aligning with the growing season of local spring maize. July recorded the highest average temperature, ranging from 20.5 °C (2020) to 23.0 °C (2021 and 2022). Among these years, 2021 experienced the highest average rainfall (723.5 mm), while 2022 had the lowest rainfall (475.4 mm) and the highest average temperature (11.0 °C). The rainfall trends during the growing season were consistent with these annual variations (Figure 2). Planting year, nitrogen application rate, film mulching, and fertilizer type significantly affected grain yield. Additionally, significant interaction effects occurred for planting year × fertilizer type, planting year × film mulching, and nitrogen application rate × fertilizer type, but not for film mulching × nitrogen application rate × fertilizer type (Table S5).

Over four years, after film mulching (PU and PS), the average grain yield under different nitrogen application rates increased significantly by 13.69–20.40% compared to traditional flat cultivation (NU and NS). Application of slow-release fertilizers (NS and PS) increased grain yield by 6.03–14.12% compared to urea (NU and PU). Among treatments, the PS treatment produced the highest grain yields under varying nitrogen application rates (125, 225, and 325 kg ha⁻¹), with increases by 6.03–23.55%, 8.12–30.17%, and 6.46–24.14%, respectively. This indicates that the combination of slow-release fertilizer and film mulching is particularly effective for boosting yield. For PS treatment, when nitrogen was applied at 225 kg ha⁻¹, the yield decreased by 8.17% compared with 325 kg ha⁻¹ but increased by 9.52% compared to 125 kg ha⁻¹ (Figure 3).

Figure 3. Grain yield (kg ha⁻¹) under different mulching and fertilizer treatments (NU: flat cultivation with urea; NS: flat cultivation with slow-release fertilizer; PU: mulching cultivation with urea; PS: mulching cultivation with slow-release fertilizer) from 2019 to 2022 (a–d). Error bars indicate standard deviations of the mean (n = 4). Lowercase letters indicate significant differences among treatments at the same nitrogen application rate; uppercase letters indicate significant differences across all treatments and nitrogen application rates (p < 0.05). N: nitrogen application rate (125, 225, and 325 kg ha⁻¹); P: plastic film mulching or no mulching; F: different fertilizer types (urea or slow-release fertilizer). ** Significant difference at 0.01 level; * Significant difference at 0.05 level; NS: no significant difference.

3.2. Average Total Greenhouse Gas Emissions and Reactive Nitrogen Losses

Over a four-year average, total greenhouse gas (GHG_Total) emissions after film mulching increased by 6.64–14.68% compared to traditional flat cultivation under varying nitrogen application rates. However, when slow-release fertilizer was used, GHG_Total emissions decreased by 5.02–7.35% compared to urea. Compared to PU, GHG_Total emissions from PS decreased by 5.02%, 6.20%, and 6.99% under varying nitrogen application rates. A significant increase in GHG_Total was observed as nitrogen application rates increased across all treatments (NU, NS, PU, and PS), while a reduction of 24.39% in GHG_Total was noted when the nitrogen application rate was reduced to 225 kg ha⁻¹ compared to 325 kg ha⁻¹. Among all agricultural inputs, nitrogen fertilizer accounted for the largest share of GHG_Total emissions, ranging from 56.12 to 84.11%, due to its production, transportation, and application processes (Figure 4a). The use of slow-release fertilizer instead of urea reduced GHG_Total emissions associated with nitrogen fertilizers by 90.54–592.72 kg CO₂-eq ha⁻¹ under varying nitrogen application rates over the four years.

Figure 4. (a) Average total greenhouse gas emissions (GHG_Total, kg CO₂-eq.kg⁻¹ ha⁻¹) and (b) reactive nitrogen (RN) losses due to NH₄⁺-N leaching (N-NH₄⁺, kg N-eq ha⁻¹), NO₃⁻-N leaching (N-NO₃⁻, kg N-eq ha⁻¹), NH₃ volatilization (N-NH₃, kg N-eq ha⁻¹), N₂O emissions (N-N₂O, kg N-eq ha⁻¹), agricultural inputs (N-inputs, kg N-eq ha⁻¹), and total RN losses (RN_Loss, kg N-eq ha⁻¹) under different mulching and fertilizer treatments (NU: flat cultivation with urea; NS: flat cultivation with slow-release fertilizer; PU: mulching cultivation with urea; PS: mulching cultivation with slow-release fertilizer) from 2019 to 2022. Error bars indicate standard deviations of the mean (n = 4). Lowercase letters indicate significant differences among treatments at the same nitrogen application rate; uppercase letters indicate significant differences across all treatments and nitrogen application rates (p < 0.05).

Reactive nitrogen losses also varied with nitrogen application rates. Over a four-year average, film mulching increased RN losses by 0.97–13.40% compared to flat cultivation, but decreased them by 27.50–27.81% when slow-release fertilizer was used instead of urea. Compared with PU, the RN losses of PS decreased by 27.50%, 27.65%, and 27.71% under varying nitrogen application rates of 125, 225, and 325 kg ha⁻¹. As nitrogen application rates increased, the RN losses rose significantly across all treatments. Specifically, at 225 kg ha⁻¹, the RN losses of PS were reduced by 30.58% compared to 325 kg ha⁻¹ (Figure 4b). The study also revealed that nitrogen leaching and ammonia volatilization accounted for 89.55–91.50% and 6.68–8.09% of total RN losses, respectively. Among nitrogen-leaching components, NH₄⁺ contributed the largest amount, while ammonia volatilization was second only to NO₃⁻ leaching. Compared to urea, slow-release fertilizer reduced the total nitrogen leaching and ammonia volatilization by 43.36–112.74 kg N ha⁻¹ year⁻¹ (Figure 4b).

3.3. Carbon Footprint and Nitrogen Footprint

Planting years, nitrogen application rates, film mulching, fertilizer types, and interactions significantly influenced the carbon footprint (CF_Yield) (p < 0.05) (Table S5). Over a four-year average, the CF_Yield of PU at a nitrogen application rate of 125 kg ha⁻¹ was the lowest, showing no significant difference from PS. At 225 kg ha⁻¹, the CF_Yield of PU was the highest, while the CF_Yield of PS was significantly reduced by 23.86% compared to PU. At 325 kg ha⁻¹, PS exhibited the lowest CF_Yield, with no significant difference from NS. Notably, CF_Yield increased significantly with higher nitrogen application rates. At 325 kg ha⁻¹, the CF_Yield rose by 38.67–110.04%, 36.95–81.70%, 29.01–206.81%, and 13.27–90.55% in the NU, NS, PU, and PS treatments, respectively, compared to 125 and 225 kg ha⁻¹. For PS, reducing the nitrogen application rate from 325 kg ha⁻¹ to 225 kg ha⁻¹ decreased CF_Yield by 14.06% (Figure 5).

Figure 5. Carbon footprint (CF_Yield, kg CO₂-eq.kg⁻¹) for various mulching and fertilizer treatments (NU: flat cultivation with urea; NS: flat cultivation with slow-release fertilizer; PU: mulching cultivation with urea; PS: mulching cultivation with slow-release fertilizer) from 2019 to 2022 (a–d). Error bars indicate standard deviations of the mean (n = 4). Lowercase letters indicate significant differences among treatments at the same nitrogen application rate; uppercase letters indicate significant differences across all treatments and nitrogen application rates (p < 0.05). N: nitrogen application rate (125, 225, and 325 kg ha⁻¹); P: plastic film mulching or no mulching; F: different fertilizer types (urea or slow-release fertilizer). ** Significant difference at 0.01 level; * Significant difference at 0.05 level; NS: no significant difference.

Similarly, nitrogen footprint (NF_Yield) was significantly (p < 0.05) affected by planting years, different nitrogen application rates, film mulching, fertilizer types, and their interactions (Table S5). Over four years, PS had the lowest NF_Yield across nitrogen application rates, showing no significant difference from NS. Compared to PU, PS reduced NF_Yield by 31.59%, 32.99%, and 32.17% under 125, 225, and 325 kg ha⁻¹, respectively. NF_Yield increased significantly with an increase in nitrogen application rates, with 325 kg ha⁻¹ showing a sharp increase of 25.30–119.61%, 30.58–110.60%, 30.31–117.10%, and 31.90–115.24% in the NU, NS, PU, and PS treatments. For PS, reducing the nitrogen application rate from 325 kg ha⁻¹ to 225 kg ha⁻¹ decreased NF_Yield by 24.18% (Figure 6).

Figure 6. Nitrogen footprint (NF_Yield, kg N Mg⁻¹) for various mulching and fertilizer treatments (NU: flat cultivation with urea; NS: flat cultivation with slow-release fertilizer; PU: mulching cultivation with urea; PS: mulching cultivation with slow-release fertilizer) from 2019 to 2022 (a–d). Error bars indicate standard deviations of the mean (n = 4). Lowercase letters indicate significant differences among treatments at the same nitrogen application rate; uppercase letters indicate significant differences across all treatments and nitrogen application rates (p < 0.05). N: nitrogen application rate (125, 225, and 325 kg ha⁻¹); P: plastic film mulching or no mulching; F: different fertilizer types (urea or slow-release fertilizer). ** Significant difference at 0.01 level; * Significant difference at 0.05 level; NS: no significant difference.

3.4. Production Cost and Net Ecosystem Economic Benefits

Planting years, nitrogen application rates, film mulching, and fertilizer types significantly impacted NEEB, CF_NEEB, and NF_NEEB (Table S5). Over a four-year average, the total cost of PS, including agricultural carbon emissions, was reduced by 8.49–19.46% compared to PU under varying nitrogen application rates. Reducing the nitrogen application rate from 325 kg ha⁻¹ to 225 kg ha⁻¹ lowered the total cost by 10.29%. The NEEB for PU and PS increased by 12.80–27.02% compared to NU and NS, and by 8.24–20.90% compared to NU and PU when slow-release fertilizer was used (NS and PS) under 125, 225, 325 kg ha⁻¹. Furthermore, increasing nitrogen application rates increased NEEB across treatments (Figure 7a).

Figure 7. (a) Total costs from fertilizers and film inputs (Total cost, CNY ha⁻¹ year⁻¹), greenhouse gas-related costs (GHG cost, CNY ha⁻¹ year⁻¹), economic returns from grain yield (GY return, CNY ha⁻¹ year⁻¹), and net ecosystem economic benefit (NEEB, CNY ha⁻¹ year⁻¹), and (b) carbon footprint per net return (kg CO₂-eq. CNY⁻¹ year⁻¹) and nitrogen footprint per net return (kg N CNY⁻¹ year⁻¹) under different mulching and fertilizer treatments (NU: flat cultivation with urea; NS: flat cultivation with slow-release fertilizer; PU: mulching cultivation with urea; PS: mulching cultivation with slow-release fertilizer) from 2019 to 2022. Error bars indicate standard deviations of the mean (n = 4). Lowercase letters indicate significant differences across nitrogen application rates for each treatment (p < 0.05).

Combining NEEB with GHG_Net emissions and RN losses, the CF_NEEB of each treatment at 125 kg ha⁻¹ did not significantly differ, but the NF_NEEB declined by 41.10%, 12.24%, and 33.85%, respectively, under PS compared to NU, NS, and PS. Compared to PU, under 225 kg ha⁻¹ and 325 kg ha⁻¹, the CF_NEEB for PS significantly declined by 27.86% and 37.28%, respectively, and the NF_NEEB reduced by 35.45% and 36.05%, respectively. The CF_NEEB and NF_NEEB increased significantly with increasing nitrogen application rate. However, reducing the nitrogen application rate from 325 kg ha⁻¹ to 225 kg ha⁻¹ decreased CF_NEEB and NF_NEEB by 14.36% and 25.17%, respectively (Figure 7b).

3.5. Energy Flow and Energy Use Efficiency

Among the various sources of energy input, fertilizer contributed the largest share, accounting for 51.48–81.61%, followed by diesel at 12.64–31.99%. Most of the energy input in the treatment came from non-renewable energy, predominantly indirect energy. Increases in nitrogen application rates, film mulching, and the use of slow-release fertilizer further elevated the proportion of indirect non-renewable energy consumption. The net energy output was highest for PS under 125, 225, and 325 kg ha⁻¹, increasing by 17.83–50.23% compared to other treatments. Net energy output also increased with higher nitrogen rates. In terms of EUE, a gradual decline was observed across treatments as nitrogen application rates increased. NU and NS had no significant difference between 225 kg ha⁻¹ and 325 kg ha⁻¹, but were notably less efficient than at 125 kg ha⁻¹. The EUE of PU and PS at 225 kg ha⁻¹ performed significantly better than at 325 kg ha⁻¹ (by 17.31% and 19.81%, respectively), and PS achieved 2.58% higher EUE than PU (Figure 8).

Figure 8. Average energy flow of input energy (Ie, 10⁻³ MJ ha⁻¹) and energy use efficiency (EUE, kg MJ⁻¹) for various mulching and fertilizer treatments (NU: flat cultivation with urea; NS: flat cultivation with slow-release fertilizer; PU: mulching cultivation with urea; PS: mulching cultivation with slow-release fertilizer) from 2019 to 2022. Error bars indicate standard deviations of the mean (n = 4). Lowercase letters indicate significant differences across different nitrogen application rates for each treatment (p < 0.05).

4. Discussion

4.1. Effect of Film Mulching and Nitrogen Fertilizer Management on Maize Yield

Consistent with its established role in soil water conservation and growth promotion [43], transparent film flat cultivation significantly enhanced grain yield by 13.69–20.40% relative to traditional methods (Figure 3). However, limited rainfall during the early growing season constrains nitrogen uptake, while nitrogen demand increases during later stages. Conventional urea applications fail to meet these dynamic needs [44,45]. Moreover, previous research has indicated that fertilizer application can significantly enhance the diversity and richness of soil bacteria, thereby fostering a more robust microbial community to support crop growth [46]. In our study, although microbial dynamics were not directly measured, it is inferred that the controlled-release mechanism of slow-release fertilizers likely provided a stable nutrient environment that favors microbial proliferation, aligning better with crop uptake patterns and significantly enhancing maize yields [47,48]. This enhancement may be further attributed to the reduction of nitrogen losses via ammonia volatilization and nitrogen leaching while potentially improving enzymatic activity related to carbon and nitrogen metabolism. These benefits are suggested to enhance net photosynthesis, increasing biomass and nitrogen accumulation [49,50,51].

Previous studies have also shown that higher nitrogen rates from slow-release fertilizers can increase grain number and weight [48,52], consistent with our findings (Table S6). However, additional N inputs no longer significantly enhance biomass or grain number, as the photosynthetic machinery and sink capacity become limited by other factors rather than N availability [22,48]. Consequently, the diminishing marginal returns at higher N rates indicate that the extra fertilizer primarily accumulates in the soil profile instead of contributing to grain formation. Furthermore, despite these productivity gains, the long-term sustainability of film mulching remains a concern. Beyond the immediate hydrothermal benefits, the accumulation of plastic film residues in the soil can lead to degraded soil structure and hindered root penetration, potentially offsetting yield benefits over time [13,16].

4.2. Effect of Film Mulching and Nitrogen Fertilizer Management on Total Greenhouse Gas Emissions and Carbon Footprint

In this study, GHG_Total emissions from mulched maize fields with optimal nitrogen fertilizer ranged from 2107.99 to 5687.52 kg CO₂-eq ha⁻¹ year⁻¹ (Figure 4a), while the CF_Yield ranged from 0.13 to 0.47 kg CO₂-eq kg⁻¹ (Figure 5). These values are lower than those reported for rainfed maize in the same region [53,54], highlighting the potential of nitrogen fertilizer management to mitigate GHG_Total emissions from mulched maize fields in the Loess Plateau region.

Nitrogen fertilizer production and application contributed the most to total emissions (Figure 4a), as reported in other studies [39,55], reinforcing the importance of optimizing nitrogen inputs. In our study, we observed that slow-release fertilizer can effectively reduce emissions under film mulching, particularly when nitrogen application is optimized [26,56]. Our results further showed that slow-release fertilizer improved soil organic matter, total nitrogen [57], and soil organic carbon (SOC) storage compared to urea [8,21,58]. This improvement is likely due to their controlled release supporting higher biomass production; specifically, it can be inferred that greater root biomass contributes to more root residues to the soil, enhancing SOC [59]. In addition, slow-release fertilizers likely create favorable conditions for microbial activity, further stabilizing organic matter and promoting SOC accumulation [60,61]. While we observed a tradeoff where higher nitrogen inputs—while increasing SOC—also resulted in higher GHG emissions due to production-related emissions of fertilizer [55,62]. Our data supported that SOC sequestration under PS at 225 kg ha⁻¹ was 57.19% lower than at 325 kg ha⁻¹, but GHG_Total emissions were also lower (Figure 4a), demonstrating the importance of balancing soil health improvements with emission reductions [57,63].

4.3. Effect of Film Mulching and Nitrogen Fertilizer Management on Reactive Nitrogen Losses and Nitrogen Footprint

Reducing nitrogen losses is crucial for minimizing environmental and human health risks [64]. Our results show that using slow-release fertilizer reduced RN losses across all nitrogen rates, with PS treatments exhibiting the lowest NF_Yield (Figure 4b and Figure 6). This improvement is attributed to the delayed hydrolysis of nitrogen in slow-release fertilizer, which curtails rapid nutrient losses [63,65]. Compared to urea, slow-release fertilizer reduced nitrogen leaching and ammonia volatilization in our study (Figure 4b), consistent with previous findings [66]. The gradual nutrient release from slow-release fertilizer reduces soil nitrogen concentrations, minimizing ammonia volatilization [67], suppresses nitrification and denitrification rates, lowering N₂O emissions [68], and limits nitrogen leaching by synchronizing release with crop demand. Additionally, the granule structure and coatings of slow-release fertilizers help further reduce leaching losses [69].

Furthermore, replacing urea with slow-release fertilizers has been shown to mitigate nitrogen leaching and ammonia volatilization while maintaining crop requirements at reduced application rates [23,70]. Beyond specialized formulations, broader precision nutrient management strategies—such as site-specific diagnostic tools and tailored fertilizers—are essential for synchronizing nutrient supply with crop demand [71]. By optimizing these inputs, nitrogen losses are minimized, directly addressing the primary driver of agricultural N₂O emissions [72,73]. Our study also found that RN losses and NF_Yield increased as the nitrogen application rate rose (Figure 4b and Figure 6). This aligns with other research showing that ammonia volatilization, N₂O emission, and N leaching are often linearly or exponentially correlated with nitrogen fertilization rates, providing valuable insights for determining optimal nitrogen application rates [74,75]. While sustained high N inputs—even via slow-release fertilizers—risk soil acidification and nitrate leaching [63], reducing the application rate from 325 to 225 kg ha⁻¹ in this study lowered nitrogen loss and footprint by 30.58% and 24.18%, respectively (Figure 4b and Figure 6). Such reductions are essential for mitigating the ecological impact of intensive agriculture and advancing sustainable intensification.

4.4. Effect of Film Mulching and Nitrogen Fertilizer Management on Net Ecosystem Economic Benefits and Energy Use Efficiency

Film mulching practices have been shown to increase ecological and economic costs associated with carbon emissions compared to flat cultivation [76]. Furthermore, factors such as labor costs, technological access, and market price influence the feasibility of slow-release fertilizers adoption. Reducing fertilizer inputs to minimize GHG_Total emissions and RN losses is only feasible if economic returns are maintained [77]. In this study, we found that slow-release fertilizer under film mulching can reduce emissions and improve economic returns (Figure 4a and Figure 7a). In addition, this approach not only helped sustain yields but also reduced energy waste, demonstrating the potential to produce more food with less energy input (Figure 8 and Figure 9). However, the holistic environmental cost should also account for the ecological risks of plastic residues. While film mulching improves energy use efficiency, the potential for long-term soil degradation and microplastic pollution necessitates the exploration of biodegradable films to ensure the system’s future sustainability [13,76].

Figure 9. Radar chart showing the effects of different mulching and fertilizer treatments. (a) NU: flat cultivation with urea; (b) NS: flat cultivation with slow-release fertilizer; (c) PU: mulching cultivation with urea; (d) PS: mulching cultivation with slow-release fertilizer on average maize yield (GY, kg ha⁻¹), average net ecosystem economic benefit (NEEB, CNY ha⁻¹ year⁻¹), average GHG_Net emissions per unit of NEEB (CF_NEEB, kg CO₂-eq. CNY⁻¹ year⁻¹), average reactive nitrogen losses per unit of NEEB (NF_NEEB, kg N CNY⁻¹ year⁻¹), average net output energy (Ne, 10⁻³ MJ ha⁻¹), and average energy use efficiency (EUE, kg MJ⁻¹) from 2019 to 2022.

While net economic and ecological benefit effectively quantifies sustainability trade-offs, these results are sensitive to market volatility and regional policy frameworks. Our findings—anchored in current local input costs and grain prices—remain subject to fluctuations in fertilizer premiums and commodity markets. Furthermore, the integration of government subsidies or carbon pricing could significantly shift the cost–benefit equilibrium. Consequently, while mulching with slow-release fertilizer at 225 kg ha⁻¹ provides a robust benchmark for the Loess Plateau, these economic outcomes should be interpreted as dynamic, scenario-dependent indicators rather than fixed parameters.

4.5. Limitations

Several uncertainties warrant consideration. First, life-cycle assessment results depend on emission factors, which may vary across years and climatic conditions. Second, long-term impacts of plastic film residues on soil structure, microbial communities, and ecosystem functioning were not quantified. Third, these findings are specific to the semi-arid soils of the study region; extrapolation to regions with different rainfall regimes or soil textures should be undertaken cautiously. Future research should prioritize long-term monitoring of SOC dynamics, in situ GHG flux measurements, and evaluation of biodegradable mulching alternatives to ensure that productivity gains do not compromise soil resilience.

5. Conclusions

This study demonstrates that integrating plastic film mulching with optimized nitrogen management can substantially improve the environmental and economic performance of rainfed maize systems in the Loess Plateau. Replacing conventional urea with slow-release fertilizer at 225 kg N ha⁻¹ achieved the most favorable balance among yield, greenhouse gas mitigation, reactive nitrogen reduction, energy efficiency, and net ecosystem economic benefit. Increasing nitrogen to 325 kg ha⁻¹ provided limited additional yield but significantly elevated carbon and nitrogen footprints, highlighting the inefficiency of high-input strategies. Future research should focus on overcoming current measurement limitations by developing robust monitoring technologies to quantify soil carbon sequestration and respiration fluxes more accurately, thereby enhancing the precision of environmental assessments in agroecosystems.