The field experiment was conducted in Daxing District (116°28′32″E, 39°58′26″N) in Beijing, North China. The experimental site has a semi-humid temperate continental monsoon climate, with an average annual temperature of 11.9°C, an average annual rainfall of 680 mm, and an altitude of 20.2 m above sea level. The soil is classified as a sandy loam (20% clay, 29% silt, and 52% sand). At the initiation of the field experiment, the following soil properties were determined in homogenized topsoil samples (0–20 cm): soil pH 8.10, electrical conductivity (EC) 75.61 μs cm⁻¹, organic matter 10.32 g kg⁻¹, total nitrogen 0.84 g kg⁻¹, NO₃⁻-N 6.98 mg kg⁻¹, available phosphorus (AP) 38.01 mg kg⁻¹, and available potassium (AK) 92.74 mg kg⁻¹. The detection methods for the soil physicochemical properties were described by Zhang et al. (2016).

Five fertilizer treatments were applied, namely: no-fertilizer treatment (control, CK), synthetic fertilizer (SF), organic N fertilizer replacing 30% synthetic N fertilizer (OF), synthetic fertilizer with 15.0 t hm⁻² zeolite added (ZSF), and organic N fertilizer replacing 30% synthetic N with zeolite added (ZOF) (details presented in Table 1). For the synthetic fertilizer (FTL-30-12-15‌‌, Futulai compound fertilizer, Beijing Futulai Compound Fertilizer Co., Ltd., Beijing, China), a compound N: P: K fertilizer (30,12,15) was applied according to local practices as basal fertilizer. The organic fertilizer (FERT-2021-001, Yite organic fertilizer) used in this experiment was commercially produced and sold by Beijing Yite Organic Fertilizer Co., Ltd. (Beijing, China). It was mainly derived from cow manure. The composition of the organic fertilizer consisted of nitrogen (N, 3.3%), phosphorus (P, 1.0%), and potassium (K, 0.7%) and had an organic matter content exceeding 64%. The zeolite (NY-FZ-200, Lingshou Zeolite), sourced from Baidu Zeolite Factory (Shijiazhuang, Heibei, China), had a mean particle diameter of 35 μm.

The five treatments were randomly arranged, and each was replicated four times. Each plot size was 40 m² (10 m × 4 m) and was separated from adjacent plots by a 1.0 m-wide buffer row (Supplementary Figure S1). All treatments were designed to have an equal N input of 300 kg N hm⁻². All the organic fertilizer, zeolite, and a portion of the synthetic fertilizer were applied as a base fertilizer in the seed bed. They were evenly spread over the field surface and then incorporated into the topsoil (0–20 cm) before planting. The maize (Zea mays L.) variety employed in the experiment was ‘Jingke 996’, with a planting density of 66,500 plants hm⁻² and a row spacing of 60 cm. The sowing date of maize was set on May 22, and the harvesting date was fixed on September 18 in 2021, 2022, and 2023. When the maize reached the stage of full expansion of the 13th leaf, the remaining N (urea, 46%, ‌‌ F-UR-46-202102) and K (KSO₄, 52%, F-K-52-202101) fertilizers were applied by top-dressing, which were produced and sold by Qian’an Fertilizer Co., Ltd. (Tangshan, Heibei, China).

All data analyses were conducted using IBM SPSS Statistics 24.0 software (SPSS Inc., Armonk, NY, USA). The means and standard errors (SE) for each treatment group were calculated. One-way analysis of variance (ANOVA, p < 0.05) was employed to detect significant differences in soil physicochemical properties, enzyme activities, microbial biomass, flux and cumulative of CO₂ and N₂O emission, GWP, and GHGI among fertilization treatments, followed by Tukey’s post-hoc test. Different lowercase letters represent significant differences between different treatments (p < 0.05). Two-way ANOVA was performed to examine the significant effects of year, and treatment, their interaction on maize yield and biomass, emission flux of CO₂ and N₂O, GWP, and GHGI. Figures were generated using Origin 2024 software (OriginLab, Northampton, MA, USA).

The kernel yield of the control (CK) decreased year by year, since no fertilizer was applied to the CK group. Compared with CK, the kernel yield (Figure 2A), aboveground biomass (AGB) (Figure 2B) and belowground biomass (BGB) (Figure 2C) of maize were all significantly increased in response to fertilizer treatment (p < 0.05) (Figure 2), although the differences in these factors across different years were not significant (Supplementary Table S1). The ZOF treatment achieved the highest yield (Figure 2A) and AGB (Figure 2B), significantly greater than those in SF by 14.72–23.61% (p < 0.05) over the 3 years, and 8.63–13.33% (p < 0.05) in 2022 and 2023, respectively. Compared with SF, the yield was significantly higher by 13.91–15.59% (p < 0.05) in ZSF in 2022 and 2023, and by 16.92% (p < 0.05) in OF in 2023. Therefore, co-applying organic fertilizer and zeolite (ZOF) demonstrated a more favorable effect on the yield and AGB of maize than did inorganic fertilizer.

From May 20 to June 15, the dynamics of soil CO₂ emission flux under different treatments exhibited a pattern of initially increasing and then decreasing. Over the 3-year trial period, the OF treatment achieved the highest peak of all the treatments, ranging from 385.30 to 393.54 mg m⁻² h⁻¹, on either June 3 or 4 (Figures 3A–C). Subsequently, the soil CO₂ emission flux gradually decreased. Notably, after fertilizer application or rainfall, the CO₂ emission flux showed a significant upward surge and peaked 4–6 days later. Compared with the CK, the average soil CO₂ emission flux (Figures 3D–F) and cumulative CO₂ emissions (Figures 3G–I) were significantly increased (p < 0.05) by the fertilizer treatments. Among these treatments, the OF treatment had the highest average CO₂ emission flux and cumulative CO₂ emissions, while the ZSF treatment had the lowest. Specifically, compared with the ZSF treatment, the cumulative CO₂ emissions from the OF treatment were significantly higher by 4.52% (p < 0.05) in 2023. The treatment had a significant impact on soil CO₂ emissions but the year had no significant effect (Supplementary Table S1). Overall, compared with the SF treatment, the OF treatment slightly increased soil CO₂ emissions.

Over the 3-year trial period, the first peak of the soil N₂O emission flux, which ranged from 448.92 to 458.02 μg m⁻² h⁻¹, was reached on May 29 in the SF treatment; the second peak, ranging from 502.91 to 529.21 μg m⁻² h⁻¹, occurred within 2–4 days after fertilizer application or rainfall in the SF treatment (Figures 4A–C). Compared with CK, the average soil N₂O emission flux (Figures 4D–F) and cumulative N₂O emissions (Figures 4G–I) were significantly higher (p < 0.05) in the fertilizer treatments. Compared with SF, the average N₂O emission flux in the OF, ZSF, and ZOF treatments were significantly lower by 6.74–8.23%, 6.10–8.79%, and 9.29–11.86% (p < 0.05), respectively; correspondingly, the cumulative N₂O emissions were significantly lower by 9.23–10.85%, 7.59–11.24%, and 12.27–16.06% (p < 0.05), respectively, in 2022 and 2023. The treatment had a significant impact on soil N₂O emissions but the year had no significant effect (Supplementary Table S1). Overall, the ZOF treatment had a greater impact than the other treatments on reducing soil N₂O emissions.

Compared with CK, the total GWP and GWP contributed by N₂O was significantly higher (p < 0.05) in the fertilizer treatments (Table 2). Compared with SF, the GWP contributed by N₂O was significantly lower by 8.45–10.17%, 7.06–10.76%, and 10.93–15.61% (p < 0.05) in OF, ZSF, and ZOF treatments, respectively, in 2022 and 2023; additionally, the total GWP of the ZOF treatment was significantly 4.78% lower (p < 0.05) than that of SF in 2023. Over the 3-year trial period, compared with the SF treatment, the GHGI in the ZSF and ZOF treatments was significantly 6.45–15.31% lower (p < 0.05) and 14.16–21.06% lower (p < 0.05), respectively; the GHGI in the OF treatment was significantly 10.53–13.13% lower (p < 0.05) in 2022 and 2023. Neither the treatment nor the year had a significant impact on GWP, although the treatment had a significant effect on GHGI (Supplementary Table S1). Overall, the ZOF treatment than the other treatments had a more pronounced effect on reducing both GWP and GHGI.

Soil pH did not exhibit any significant difference across the various treatments (Figure 5A). Compared with the SF treatment, bulk density was significantly 4.26% (p < 0.05) lower in the ZOF treatment (Figure 5B). Compared with CK, the concentrations of soil organic carbon (Figure 5C), total nitrogen (Figure 5D), dissolved organic carbon (DOC) (Figure 5E), NH₄⁺-N (Figure 5F), NO₃⁻-N (Figure 5G), available potassium (Figure 5H), and available phosphorus (Figure 5I) were significantly higher (p < 0.05) in the fertilizer treatments. Compared with SF, soil NH₄⁺-N and NO₃⁻-N concentrations were significantly lower by 12.27, 9.08 and 10.56, 9.30% (p < 0.05) in the OF and ZOF treatments, respectively, while soil available potassium and phosphorus concentrations were significantly higher by 7.34 and 8.90% (p < 0.05) in the ZOF treatment, respectively. Compared with SF, soil DOC concentration was significantly higher by 15.84, 12.62, and 19.48% (p < 0.05) in OF, ZSF, and ZOF treatments, respectively. Overall, the ZOF treatment had the greatest impact on improving soil physicochemical properties, particularly in terms of available nutrient concentrations.

Compared with CK, the activities of soil enzymes β-glucosidase (Figure 6A) and nitrate reductase (Figure 6B), as well as the concentrations of MBC (Figure 6C) and MBN (Figure 6D), were significantly higher (p < 0.05) in the fertilizer treatments. Compared with SF treatment, the activity of soil β-glucosidase was significantly higher (p < 0.05) by 11.68 and 9.20% in OF and ZOF treatments, respectively, while the activity of nitrate reductase was significantly lower (p < 0.05) by 8.79% in the ZSF treatment. Compared with SF, MBC and MBN concentrations were significantly higher by 8.42 and 11.29% in the ZOF treatment (p < 0.05), respectively. In general, the ZOF treatment significantly increased soil β-glucosidase activities and microbial biomass in the soil.

Significant positive correlations (p < 0.05) were observed between soil CO₂ emissions and several soil parameters, including WFPS, BG activity, and soil concentrations of DOC, OM, MBN, and MBC. Similarly, soil N₂O emissions exhibited significant positive correlations (p < 0.05) with soil WFPS, NR activity, and concentrations of NO₃⁻-N, TN, MBC, and MBN. Furthermore, maize yield was significantly positively correlated (p < 0.05) with multiple soil factors such as WFPS, BG activity, and concentrations of DOC, NH₄⁺-N, AP, AK, OM, TN, MBC, and MBN. These results suggested that GHG emissions and maize yield in sandy loam soil were closely related to soil moisture, available nutrients, and microbial biomass (Figure 7).

Soil CO₂ emissions in farmland primarily originate from the heterotrophic respiration of microorganisms and autotrophic respiration of crop roots (Galic et al., 2019). In the current study, the first peak of soil CO₂ emission fluxes in each fertilizer treatment occurred on June 3 and 4. This peak was ascribed to the soil temperature and humidity becoming more suitable for microbial activity at this stage, so that the nutrients in the base fertilizer were transformed to provide sufficient nutrition for microorganisms, thereby facilitating their heterotrophic respiration. Notably, the soil CO₂ emission fluxes in the OF treatment were higher than those in the SF and ZSF treatments during this stage. This was because a large quantity of exogenous organic carbon from organic fertilizer was incorporated into the soil and directly served as the respiratory substrate for microorganisms, promoting soil CO₂ emissions (Wang et al., 2022; Chaker et al., 2023; Jaiswal et al., 2024). By mid-June, as the easily degradable organic substances in the organic fertilizer were being consumed, the soil CO₂ emission fluxes under the OF treatment decreased and were no longer significantly different from those under the SF treatment. During the middle growth period of maize, a secondary peak in soil CO₂ emission fluxes occurred in each treatment. This might be because the abundant precipitation and relatively high soil temperatures (Supplementary Figure S2) during this period enhanced the activity of soil microorganisms. Particularly after the top-dressing of urea in late July, the concentration of available N in the sandy soil increased rapidly, providing sufficient N for the growth of both plants and microorganisms. At this point, the respiration of maize roots was intense, thus accelerating the soil CO₂ emission fluxes (Oraegbunam et al., 2022; Liu et al., 2024a).

Throughout the maize growing season, compared with SF, OF slightly increased the cumulative soil CO₂ emissions, which was consistent with the research results of Jaiswal et al. (2024) and Liu et al. (2024a). The underlying mechanisms for increased CO₂ emissions under OF treatment can be attributed to three key factors. First, the part-replacement of synthetic fertilizer by organic fertilizer increased the concentration of active organic carbon (from organic matter) in the sandy loam soil. This provided an adequate concentration of substrate for microbial respiration, promoted the metabolic activities of microorganisms, and directly increased CO₂ emissions (Bhunia et al., 2021; Delucena et al., 2023). The positive correlations observed between microbial biomass carbon, soil β-glucosidase activity, and soil CO₂ emissions observed in this study supported this mechanism. Second, organic fertilizers application improved soil fertility, which in turn promoted the growth and metabolism of maize roots. As a result, the CO₂ emissions from root respiration increased. The increase in belowground biomass observed in this study also supported this hypothesis (Bhunia et al., 2021; Yan et al., 2023). Third, the introduction of exogenous organic matter into sandy loam soil accelerated the decomposition of the original soil organic matter, triggering a positive priming effect. Collectively, these mechanisms contributed to an increase in soil CO₂ emissions under the OF treatment (Chaker et al., 2023).

In the current study, compared with the SF treatment, the ZSF treatment reduced soil CO₂ emissions to a certain extent, while the ZOF treatment had an insignificant impact on soil CO₂ emissions. This phenomenon was ascribed to the following reasons. First, the application of zeolite powder with a small particle size (35 μm diameter) to the sandy loam soil increased the soil bulk density and decreased the porosity. To some extent, this led to a reduction in soil oxygen concentration, resulting in a decrease in the available oxygen within the soil. Consequently, the respiration rate of microorganisms and roots was inhibited, thereby reducing CO₂ emissions (Liu et al., 2024a). However, zeolite addition had no significant effect on soil bulk density in clay soils (Ersin et al., 2004; Obalum and Obi, 2010), moreover, the oxygen diffusion coefficient in clay soils is significantly lower than that in sandy soils, which inherently inhibits aerobic respiration. Therefore, zeolite application that reduces soil CO₂ emissions may be more suitable for sandy soils. Second, the sandy loam soil had a relatively low clay content, and the cohesion among soil particles was poor. As a result, it was challenging to form large aggregates and hence large pores to achieve soil aeration. The application of zeolite and exogenous organic matter could directly or indirectly supply a cementing agent for soil aggregates. This promoted the formation of large aggregates and enhanced the stability of soil aggregates through physical protection, thus reducing the decomposition of soil carbon (Abdalla et al., 2022; Hei et al., 2024). In addition, the application of zeolite not only decreased soil water evaporation but also increased the soil water-binding capacity (Nakhli et al., 2017). Given that CO₂ has a high solubility in water, the application of zeolite decreased the diffusion of CO₂ to the soil surface (Abdalla et al., 2022).

Because soil N₂O generation and emission are under the regulation of N fertilizer management strategies, a rationalized N fertilizer approach could effectively curtail soil N₂O emissions (Wu et al., 2024). In the present study, the peak value of soil N₂O flux occurred following N fertilizer application, with the highest peak occurring under the SF treatment (Figure 4). Application of synthetic N fertilizer typically resulted in an excess of soil inorganic N. This surplus provided abundant substrates for both nitrification and denitrification, two complementary microbial processes in agricultural soils that generate N₂O (Shu et al., 2021; Chataut et al., 2023). Our research revealed that, compared with SF, the OF treatment significantly reduced soil N₂O emissions by 9.23–10.85% (Figure 4), indicating that replacing 30% of synthetic fertilizer N with organic fertilizer N mitigated soil N₂O emissions in a sandy loam soil. This finding aligned with results from previous research in loam and clay soils, which showed that substitution with organic N had a positive impact on soil quality and could reduce N₂O emissions (Liang et al., 2024; Park et al., 2024). Several mechanisms can account for the decrease in N₂O emissions in response to organic fertilizer N substitution. First, soil NO₃⁻ and NH₄⁺, serving as key substrates and reactants in nitrification and denitrification processes, play pivotal roles in regulating N₂O fluxes (Wu et al., 2022). Organic N fertilizer part-substitution for synthetic N fertilizer reduced soil NO₃⁻ and NH₄⁺ concentrations compared with sole inorganic N fertilizer application (Figure 5), a finding which was similar to the results of Wu et al. (2024). Generally, organic fertilizers predominantly supply organic N, which is gradually converted to inorganic N through N mineralization, then the mineral N is quickly absorbed by plants (Jaiswal et al., 2024). As a result, organic substitution diminishes the supply of mineral N substrates for nitrifying and denitrifying bacteria, thereby reducing N₂O emissions. This phenomenon is further supported by the positive correlations between N₂O emissions and soil NO₃⁻ and NH₄⁺ concentrations observed in both the current study (Figure 7) and prior research (Xie et al., 2024).

Second, organic fertilizers release N slowly after the decomposition of organic matter, limiting the risk of N being converted into N₂O, N₂, or ammonia (Shu et al., 2021). Moreover, organic fertilizers can enhance the retention of available N in the soil, facilitating its uptake by plants and minimizing excessive gaseous N losses (Cheng et al., 2022; Liang et al., 2024). Finally, organic fertilizers can promote the reduction of N₂O to N₂ by enhancing electron flow in denitrification, especially in low-NO₃⁻ soils (Tang et al., 2024). In the present study, the soil NO₃⁻ concentration was relatively low in the OF treatment, which led to the reduction of N₂O emissions.

The results indicated that, compared with the SF treatment, the peak values of N₂O emission fluxes and cumulative N₂O emissions were lower in zeolite application treatments (ZSF and ZOF) than in no-zeolite treatments. This suggested that the application of zeolite mitigated soil N₂O emissions in sandy loam soils, a conclusion that aligned with research findings from paddy soils (Sha et al., 2020). Several underlying factors can account for this phenomenon. Firstly, due to its high adsorption capacity, zeolite can “dampen the peaks” of soil nutrients following N fertilizer application. It reduces the concentrations of reaction substrates (NO₃⁻ and NH₄⁺) for soil nitrification–denitrification processes following N application. As a result, zeolite mitigates the peak values of soil N₂O generation and emission after the application of both base and top-dressing N fertilizers (Ippolito et al., 2011; Liu et al., 2024a). Second, zeolite can bind NH₄⁺ within its pore structure, providing a long-lasting fertilizer effect. Moreover, the small channels within zeolite can prevent nitrifying bacteria from accessing the bound NH₄⁺, thus inhibiting the nitrification of NH₄⁺. This effectively reduces the loss of active N (Jumadi et al., 2020). Thirdly, zeolite has a high water-holding capacity. It can regulate water flow within soil pores and mitigate N₂O emissions by reducing the nitrification process in the soil (Zheng et al., 2024). Lastly, the application of zeolite can effectively suppress nitrate reductase activity (Figure 6), a key enzyme in denitrification pathways, thereby reducing N₂O emissions (Tzanakakis et al., 2021).

The application to soil of zeolite significantly mitigated soil N₂O emissions primarily through two integrated mechanisms when organic fertilizer replaced 30% of the synthetic N in sandy loam soil (Figures 4, 5). First, the N release rate from the organic fertilizer, in combination with zeolite’s capacity to adsorb part of NH₄⁺-N and NO₃⁻-N, prevented the concentrations of inorganic N in the soil from reaching excessively high levels. This phenomenon reduced the substrates available for nitrification and denitrification, thereby decreasing the peak value of N₂O emission flux and N₂O cumulative emissions in the ZOF (zeolite-organic-synthetic fertilizer) treatment (Figure 4). The increase in soil organic N, which was introduced through the application of organic fertilizer, can promote the assimilation of organic N by soil microbial activity, immobilize more N by microbial activity, and reduce the amount of N available for the production of N₂O (Nakhli et al., 2017; Cheng et al., 2022). This present study further confirmed that the ZOF treatment significantly increased the MBN concentration (Figure 6); during the maize growing season, the N₂O fluxes were significantly negatively correlated with MBN concentration (Figure 7) (Zheng et al., 2024). Second, zeolite enhanced the soil aggregate structure, while organic fertilizers increased soil organic matter and microbial activities. Together, they created a favorable soil environment for crop growth and microbial activity, increased the activity of denitrifying enzymes, enhanced the reduction of NO₃⁻ to N₂, and inhibited the production and emission of N₂O (Park et al., 2024).

Soil temperature and moisture are also pivotal factors influencing N₂O emissions (Wu et al., 2022; Jaiswal et al., 2024). N₂O emissions were substantially higher during the middle stage of the maize growth period compared with the early stage. This difference can be mainly attributed to the higher soil temperature and moisture levels in the middle stage of maize growth (Figure 1 and Supplementary Figure S2). These elevated conditions accelerate the N-cycling process, thereby enhancing N₂O production (Jumadi et al., 2020; Zhong et al., 2021). When water acts as a limiting factor, the ZOF treatment increased the N₂O emission flux during this stage, due to an increase in the WFPS content (Supplementary Figure S3). Notably, WFPS exhibited a significant positive correlation with soil N₂O emissions, as reported by Fan et al. (2024).

Soil quality is a multifaceted indicator that reflects the soil’s potential to sustain its ecological functions, which depends on the interplay of physical, chemical, and biological properties (Li et al., 2022). The present study demonstrated that the ZOF treatment augmented the concentrations of soil organic matter, dissolved organic carbon, available potassium, and available phosphorus in sandy loam soil. Additionally, it increased the activities of soil β-glucosidase and the concentrations of microbial biomass carbon and nitrogen, while reducing the soil’s bulk density. This finding is largely consistent with previous research, indicating that, when the organic fertilizer, replacing 30% of the synthetic fertilizer N, and zeolite were applied together, soil quality was improved (Park et al., 2024).

The soil treated with organic fertilizer and zeolite co-application was also more effective at retaining soil nutrients. Organic fertilizers are enriched with organic matter containing various nutrients and can provide a rich source of nutrients to the soil. The slow-release nature of nutrients from organic fertilizers ensures a continuous nutrient supply (Hei et al., 2024; Wu et al., 2024). Zeolite, with its distinctive porous structure and high cation-exchange capacity, can adsorb and store nutrient ions such as NH₄⁺, NO₃⁻, K⁺, PO₄³⁻, and Mg²⁺, preventing these nutrients from being leached away by rainwater or irrigation (Latifah et al., 2017; Baghbani-Arani et al., 2020; Ferretti et al., 2024), and minimizing NO₃⁻ losses in sandy loam soil of North China (Li et al., 2022). Consequently, the combined application of zeolite and organic fertilizer further bolsters the soil’s nutrient-retention capacity.

The co-application of organic fertilizer and zeolite can enhance the aggregation of soil particles (Lin et al., 2025). The organic colloidal substances in organic fertilizers can interact with zeolite and bind soil particles together, forming more stable aggregates. The activity of soil microorganisms is enhanced under the ZOF treatment. Organic fertilizers provide labile carbon sources and energy for microorganisms, spurring their activity and reproduction, and increasing microbial biomass carbon and nitrogen (Liu et al., 2024b). Simultaneously, during the decomposition of organic fertilizers and the utilization of nutrients adsorbed by zeolite, the metabolic activities of microorganisms increase. This leads to an increase in the activities of soil enzymes, such as β-glucosidase and nitrate reductase activity, and these enzymes play a crucial role in the transformation of soil carbon and nitrogen nutrients (Yang et al., 2019).

Stable or increased production is one of the most effective ways to evaluate agricultural practices (Wang et al., 2024). Previous studies have confirmed that the reasonable application of organic fertilizers and the addition of zeolite are important management strategy for increasing crop yields in sandy soils (Liu et al., 2024c; Zheng et al., 2024). The current study obtained similar results, demonstrating that the co-applied organic fertilizer and zeolite could significantly enhance maize yield, primarily as a result of their ability to provide sufficient nutrients and foster a suitable growth environment for maize plants. Previous studies have confirmed that rationalized application of organic fertilizers, in conjunction with the addition of zeolite, serves as a pivotal management strategy for enhancing crop yields in sandy soil (Liu et al., 2024c; Zheng et al., 2024), underscoring their applicability across temperate sandy agroecosystems.

When applied to the soil, zeolite exhibits remarkable properties. It can adsorb a substantial quantity of water and nutrient ions due to its microporous structure and high cation exchange capacity, thereby minimizing water and nutrient losses that typically occur after fertilizer application. Moreover, when the concentrations of water and nutrients in the soil are relatively low, zeolite slowly releases the adsorbed substances (Tarkalson and Ippolito, 2011; Zheng et al., 2024). Through this dynamic adsorption-release balance of zeolite, the various nutrient elements from organic fertilizers are gradually released into the soil, ensuring the continuous supply of available water and nutrients in sandy soil, mitigating the negative impacts of water and nutrient deficiency stress on maize yield, and improving water- and fertilizer-use efficiency to a certain extent (Omar et al., 2018; Liu et al., 2024c; Wang et al., 2024). In addition, a relatively well-developed root system directly augments the plant’s capacity to take up water and soil nutrients, including NH₄⁺, NO₃⁻, available phosphorus, and available potassium (Baghbani-Arani et al., 2020; Liu et al., 2023). In addition, such a robust root system effectively enhances the transport efficiency of nutrients and water from the soil to the shoots, increasing leaf relative water content by 12% and photosynthetic rate by 15%, consequently increasing nutrient accumulation and aboveground biomass (stems and leaves) and grain yield of maize (Park et al., 2024). Furthermore, appropriate concentrations of soil water and nutrients stimulate microbial biomass and shift community composition, and reduce nutrient loss by carbon and nitrogen fixation by microbes, enhancing the utilization efficiency of water and nutrients by maize and ultimately contributing to an increase in maize yield (Latifah et al., 2017; Park et al., 2024). In conclusion, the co-addition of organic fertilizer (replacing 30% of the synthetic N fertilizer) and zeolite represented an effective approach to improving maize yield in sandy loam soil and promoted the sustainable production of such farmland. However, the optimal ratio of the application of zeolite to organic fertilizer for the improvement of loamy sandy soil still requires systematic investigation.

In the current study, the GWP was calculated based on the fluxes of two GHGs, CO₂ and N₂O, expressed in terms of CO₂-equivalents, following the methodology proposed by Chataut et al. (2023). Consistent with previous research (Liu et al., 2024b; Xie et al., 2024), our findings demonstrated that the OF and ZSF treatments significantly reduced the contribution of N₂O emissions to GWP during the maize growing season (Table 2); this may be because both the replacement of 30% of synthetic N fertilizer by organic fertilizer and the application of zeolite inhibited N₂O emissions (Figure 4). Additionally, the ZOF treatment significantly decreased the total GWP in 2023. This is likely because co-application of organic fertilizer and zeolite inhibited both N₂O and CO₂ emissions. The GHGI depends on the grain yield of maize and the total greenhouse gas emissions from the soil and is a variable which can better reflect the relationship between economic and environmental benefits in agricultural systems (Zhang et al., 2019). Compared with the SF treatment, the GHGI was lower in the OF, ZSF, and ZOF treatments during the 3 years of trials (Table 2), and was due to the increase in maize yield (Figure 2) and decrease in the N₂O-related contribution to GWP. The present research indicated that the ZOF treatment had the most pronounced effect in reducing CO₂ and N₂O emissions and improving maize production by promoting soil physicochemical and biological properties in sandy loam soil, thus confirming our initial hypothesis. The organic N fertilizer replacing 30% synthetic N with 15.0 t hm⁻² zeolite added (ZOF) hold great promise for reducing GHG emissions, enhancing maize yields, and improving soil quality in the sandy loam soils of the North China Plain. Overall, our research has not only provided invaluable insights into the sustainable development of sandy loam farmlands, but also offered a technical basis for yield enhancement and GHG emission reduction (particularly N₂O) in vegetable cultivation with high fertilizer requirements in this region. However, in farmlands with other soil types in temperate regions, it is essential to adjust the application scheme of zeolite-organic amendments according to local soil types to achieve the synergistic objectives of carbon sequestration and yield enhancement, complemented by appropriate agricultural extension programs to enhance regional climate adaptability.