The experiment was carried out at the Jindong Organic Dryland Farming Experimental Demonstration Base of Shanxi Agricultural University in Dianshang Town, Huguan County, Changzhi City, Shanxi Province (E113°27′, N36°03′), within a steel-framed arched plastic greenhouse (90 m long, 8 m wide, 3.5 m high at the ridge, and 1.8 m high at both sides, with natural ventilation). This site has a warm temperate semi-humid continental monsoon climate, with an average annual temperature of 10 °C and rainfall of 574.5 mm. The experiment was conducted in a completely randomized design with four treatments: no mulching (CK), straw mulching (SM, whole straw was covered on the ridges, 650 kg·ha⁻¹), plastic film mulching (SBF), and combined film-straw mulching (SBFSF, plastic film on ridges with straw in furrows, 650 kg·ha⁻¹). Each treatment had three replications, and every plot with five beds and five furrows. Planting specifications included two rows per bed, 35 cm inter-plant spacing, 40 cm inter-row spacing, 60 cm ridge width, 80 cm furrow width, and 22 tomato plants per row. Drip irrigation lines were placed along the planting rows. Tomato seedlings were transplanted on May 22, 2024, and the vines were removed on September 15, 2024. Standard agronomic practices, such as irrigation, pruning, and pesticide application (Supplementary Table S1), were uniformly applied to all treatments. The tomato variety tested was “Rirun No. 8,” utilizing 800 mm wide, 0.01 mm thick silver-black plastic film and corn straw from the previous year’s harvest at the site for mulching. The initial soil physicochemical indicators at 0–20 cm soil layer are provided in Table 1.

The dry weight and root morphology in the 0–20 cm soil layer were monitored every 20 days post transplantation of the tomato plants. Following this, the tomato plants were brought back to laboratory, where their dry weights were quantified after 105 °C of exposure for 30 min, drying them in an oven at 80 °C until the constant weight and weighing the samples (Wang F. et al., 2021). The root system determination method according to the following procedure: a custom-designed root extractor (in a cube shape, 20 cm × 20 cm × 20 cm, 3 mm thick, with one side shaped like a knife to facilitate soil penetration) was inserted into the soil at the tomato root location and driven in using a hammer after removing the aboveground portion of the plant. Subsequently, the root-soil block was then excavated with a spade and trimmed neatly using a professional cutting tool to obtain a standardized sample. The excavated root system along with the soil was transported to the laboratory, then carefully washed off the soil adhering to the roots. The root system was then submerged in a glass tank filled with water, where the roots were carefully separated and scanned using an LA2400 root scanner (Canadian Reagent Instruments Company) to collect data on total root length and root surface area. Finally, the scanned roots were then oven-dried at 80 °C to a constant weight for calculating the dry matter weight of the root biomass.

During the harvest seasons of 2022–2024, tomatoes were harvested in batches from the two middle rows of each plot, and the yield was calculated on a per-hectare basis.

The data were organized using Microsoft Excel 2007 (Microsoft Corp., USA), and variance analysis was performed with SPSS 27.0 (SPSS Inc., Chicago, USA), employing Duncan’s method for significance testing, including tomato yield, dry matter weight, root morphology, soil physical and chemical properties as well as soil bacterial and fungal diversity index (all data were subjected to tests for normality and homogeneity of variances, and for datasets that did not meet the assumptions of normal distribution or homoscedasticity, appropriate data transformations were applied). Principal coordinates analysis (PCoA) was conducted using the Bray-Curtis distance matrix algorithm to compare the taxonomic composition among treatments. Functional prediction of microbial communities was conducted using PICRUSt2 software. Redundancy analysis (RDA) based on a linear model was performed to assess the effects of soil physical and chemical properties on the structure of soil bacterial and fungal communities. Partial least squares path model (PLS-PM) was used to explore relationships among soil physical properties, soil chemical properties, soil bacteria, soil fungi and tomato plant. Both RDA and PLS -PM analyses were implemented using the vegan and plspm packages in R language, respectively. Figures were generated using Origin 2022.

Soil chemical properties exhibited significant change under different mulching treatments after 3 years of tomato cultivation (Figure 4). Compared to the control (CK), SM and SBFSF significantly increased the contents of soil organic matter, total nitrogen, available potassium, and available phosphorus, with corresponding increases of 5.92%, 6.34%, 18.15%, and 2.60% (p < 0.01) for SM, and 4.04%, 4.46%, 10.91%, and 2.15% (p < 0.01) for SBFSF. In contrast, SBF resulted in a significant reduction in soil organic matter content by 3.85% (p < 0.01), though it increased available potassium content by 4.98% (p < 0.01). All mulching treatments significantly reduced soil pH.

SBFSF significantly increased only the fungal Chao1 index, while showing significant reductions in all other alpha-diversity indices for both bacterial and fungal communities. In contrast, SM exhibited significant increases across all measured alpha-diversity indices. Regarding SBF, no significant difference was observed in the bacterial Chao 1 index compared to CK, however, all remaining alpha-diversity indices showed statistically significant enhancement (Figures 5a–d).

The composition of bacterial and fungal communities in soil was different under various mulching treatments, as shown by the Principal Coordinates Analysis (PCoA). The first axis of variation for the bacterial community explained 23.5%, while the second axis explained 21.4% in community structure. The values for all of the mulching treatments had a clear separation from control (CK) (Figure 5e). In contrast, although the fungal communities under various mulching treatments did not overlap with CK in the ordination space, the SBFSF treatment was relatively closer to CK. The first and second axes explained 59.7 and 20.4% of the variance in fungal community composition, respectively (Figure 5f).

The composition and the relative abundance of soil bacteria were analyzed at both the phylum (top 10) and genus (top 10) levels. Compared to CK, SBFSF, SM and SBF significantly reduced the relative abundance of Proteobacteria by −17.20% to −4.60%. Compared to other treatments, SBFSF significantly increased the relative abundance of Firmicutes by +125.18 to 193.55%, while SM enhanced Actinobacteria (+2.63% to +22.04%) and Chloroflexi (+50.70–71.87%). SBF showed a significant increase in Acidobacteria (+21.79–37.12%) (Figure 6a). At the genus level, SBFSF enriched Lysobacter (+49.19% to +186.62%) and Bacillus (+168.56% to +273.00%), whereas reduced the relative abundance of Sphingomonas (−28.32% to −9.60%). SM increased JG30-KF-CM45 (+59.61% to +73.40%) and Arthrobacter (+28.94% to +55.42%), whereas reduced the relative abundance of Sphingomonas (−46.51% to −25.37%). SBF led to higher relative abundance of Vicinamibacteraceae (+19.34% to +41.39%) (Figure 6b).

Fungal community analysis revealed that Ascomycota was the dominant phylum with relative abundances ranging from 86.76 to 94.38% (Figure 6c). Compared to other treatments, SBFSF significantly increased the relative abundances of Ascomycota (+7.99% to +8.19%) and Mortierellomycota (+11.05% to +98.71%). At the genus level, the relative abundance of Trichocladium in SBFSF was significantly higher than that in other treatments, showing an increase of 30.43 to 269.95%, while SM treatment notably enriched Fusarium (+120.39% to +149.16%), Cladosporium (587.22% to +4141.65%), and Alternaria (10953.15% to +32375.165). Moreover, SBF displayed significant increases in both Cephaliophora (+218.39% to +454.82%) and Fusarium (+12.48% to +180.25%) (Figure 6d).

Distinct soil microbial taxa enriched in various mulching treatments were identified through LEfSe analysis (Supplementary Figure S2). Overall, the numbers of indicator bacterial taxa associated decreased across all mulching types. Specifically, the SM and SBFSF management strategies promoted the enrichment of fungal taxa, while the SBF practice had a diminishing effect. Notably, the bacterial genus Lysobacter showed enrichment in response to SBFSF treatment, while Chloroflexi predominated under SM and Acidobacteriota under SBF. Among fungal taxa, Trichocladium was enriched in SBFSF, Dothideomycetes in SM, and Microascales in SBF.

Functional prediction of bacterial metabolism revealed distinct metabolic profiles in SBFSF compared to SM and SBF, with enhanced activities in amino acid metabolism, carbohydrate metabolism, metabolism of other amino acids, and lipid metabolism (Figure 6E). For fungal metabolic functions, SBFSF demonstrated increased capabilities in multiple pathways relative to SM and SBF, including electron transfer, respiration, secondary metabolite biosynthesis, cofactor and prosthetic group biosynthesis, electron carrier synthesis, vitamin biosynthesis, fermentation, as well as acid and lipid biosynthesis (Figure 6F).

At the bacterial genus level, Bacillus showed a significant positive correlation with metabolism of other amino acids, with a notably high correlation coefficient of 0.83 for carbohydrate metabolismz although not statistically significant. Lysobacter exhibited strong correlations with amino acid metabolism (0.85), lipid metabolism (0.76), and metabolism of other amino acids (0.86). Additionally, 67–14 showed a significant positive correlation with carbohydrate metabolism and a notably high correlation coefficient of 0.70 for other amino acid metabolism (Figure 6G). At the fungal genus level, Mycochlamys demonstrated substantial correlations with electron carrier and vitamin biosynthesis (0.92), electron transfer (0.84), respiration (0.84), and carbohydrate biosynthesis (0.88) (Figure 6H).

Redundancy analysis conducted at the bacterial genus level, focusing on the top five scoring bacteria, revealed that the two principal axes explained 49.2 and 36.2% of the total variance contribution, respectively (Figure 7a). Specifically, the application of SBFSF was found to enhance the proliferation of Bacillus and Lysobacter. Bacillus exhibited a significant positive correlation with soil water content (SWC) (p < 0.001) and a negative correlation with pH (p < 0.05). Moreover, the use of SM was observed to stimulate the growth of JG30-KF-CM45, which displayed a negative correlation with soil bulk density (SBD) (p < 0.05). Additionally, Sphingomonas showed negative correlations with available potassium (AK), available phosphorus (AP), total nitrogen (TN), and soil organic matter (SOM) (p < 0.01). Similarly, redundancy analysis conducted at the fungal genus level, focusing on the top five fungi, in relation to soil environmental factors, indicated that the two principal axes explained 48.8 and 29.1% of the total variance, respectively (Figure 7b). The application of SBFSF was found to promote the growth of Trichocladium, while SM significantly enhanced the growth of Alternaria. Alternaria exhibited a positive correlation with total nitrogen (TN) (p < 0.05) and a negative correlation with soil bulk density (SBD) (p < 0.01). Furthermore, the use of SBF was associated with increased growth of Cephalophora, which displayed negative correlations with soil organic matter (SOM), total nitrogen (TN), and available potassium (AK) (p < 0.05).

Further analysis using partial least squares path modeling (PLS-PM) revealed that a combination of soil bacterial community composition, diversity, soil properties, and plant growth accounted for 60.3% of the variability in tomato yield (Figure 7c). Similarly, soil fungal community composition, diversity, soil properties, and plant growth elucidated 60.8% of the variance in tomato yield (Figure 7d). Mulching practices exerted a direct and significant positive effect on soil bacterial community composition (path coefficient = 0.578, p < 0.010) and diversity (path coefficient = 0.599, p < 0.010), but only bacterial community composition had significant negative impacts on soil properties (path coefficient = −0.804, p < 0.001), which subsequently positively influenced tomato plant growth (path coefficient = 1.375, p < 0.001) and ultimately enhanced tomato yield (path coefficient = 0.923, p < 0.001). The mulching practices exerted a direct and positive effect on the composition of the soil fungal community (path coefficient = 0.806, p < 0.001), consequently positively influenced soil properties (path coefficient = 0.406, p < 0.050). Conversely, it negatively affected fungal diversity (path coefficient = −0.708, p < 0.001), which subsequently negatively affected soil properties (path coefficient = −0.417, p < 0.050). The improved soil properties then positively promoted tomato plant growth (path coefficient = 0.870, p < 0.001), ultimately benefiting tomato plant growth (path coefficient = 0.924, p < 0.001).

Previous studies has shown that both plastic film mulching and straw mulching are both able to prevent water movement and atmospheric gaseous exchange in soil (Chai et al., 2022; Zhang et al., 2021; Zribi et al., 2015), thus enhancing soil hydrothermal properties compared with bare plots. Our study found that all three mulching treatments enhanced soil water retention capacity, with SBFSF exhibiting the highest moisture retention due to the complementary functions of its components within a fully covered microenvironment. At 80 days after transplantation, no significant differences in soil water content were detected among the three mulching treatments. This result may be explained by the high transpiration rates of tomato plants under all mulching practices during this stage, leading to increased soil water depletion and consequently diminished variation in soil moisture levels.

Mulching practices enhanced soil temperature, with increases ranging between 0.01–2.21 °C at 10 cm depth and 0.02–2.05 °C at 20 cm depth, which is in agreement with former study results (Miao et al., 2025; Wang Y. et al., 2025; Qu and Feng, 2022). In this study, both SBFSF and SM demonstrated the cooling effect under high temperatures and warming effect under low temperatures, indicating that the soil temperature regulation effects of different mulching practices exhibited temporal variations. SBF exhibited the strongest warming effect in the early stage of tomato growth, but it could not regulate temperature during high-temperature period. In contrast, SBFSF maintained the most stable soil thermal conditions during the entire growth period, which implied combined effect of full soil coverage on decreasing temperature fluctuation. These results emphasized the superior performance of SBFSF in generating a balanced and flexible thermal environment and such characteristics might contribute to enhancing tomato growth and yield stability under dynamic climate.

Our study demonstrated that both SM and SBFSF enhanced rhizosphere soil fertility after 3 years of tomato cultivation by increasing the content of soil organic matter, total nitrogen, and available nutrients (Zhu et al., 2022). This improvement occurred because straw mulching on the soil surface releases nutrients through microbial decomposition (Liu et al., 2014), thereby enriching the topsoil nutrient pool—a finding consistent with the research of Biswal et al. (2022), Wang J. et al. (2021), and Chen et al. (2022). SM treatment reduced the relative abundance of Sphingomonas, and RDA analysis indicated that Sphingomonas was negatively correlated with soil available potassium, available phosphorus, and organic matter (p < 0.01). Therefore, it can be inferred that Sphingomonas played an important role in the soil fertility of SM treatment. The improvement in soil fertility under SBFSF was also associated with its regulation on soil temperature. Small fluctuation of soil temperature in return leads to the low mineralization rates of SOC and available nutrients (Zhou et al., 2012). We attribute the increased soil fertility under SBFSF to the combined regulatory effects of plastic film on ridges and straw in furrows. SBFSF likely produces a more favorable soil microenvironment which together support improved soil fertility. In addition, SM reduced soil bulk density which contributed to crop growth and roots residue input into soil for conversion into plant available nitrogen and phosphorous nutrient (Wang et al., 2024). Under SBF, initially higher soil temperatures enhance the process of organic matter decomposition and exhaustion of available nutrient resources at a faster rate, leading to lower rhizosphere fertility.

Soil microorganisms are the most dynamic components of soil, with their community diversity and composition being influenced by mulching practices. In the present study, all mulching treatments significantly affected the richness and diversity of rhizosphere soil bacterial and fungal populations, and was due to the difference in soil temperature, moisture, and organic matter input by mulching practices (Luo et al., 2019; Ling et al., 2016; Chen et al., 2017). SBFSF significantly increased fungal Chao1 index and decreased all other alpha-diversity indices for both bacteria and fungi. This was an indication that although SBFSF could promote the richness of fungus, it might also act as a limited ecological niche and inhibit overall microbial diversity (Zhao et al., 2024). On the contrary, SM caused significant increases in all alpha-diversity measures which may indicate that a niche is less selective for richness and evenness. A majority of the previous reports suggested that straw mulching increased bacterial diversity (Zhang et al., 2024), while results on fungal communities have not been consistent (Fu et al., 2019; Liu et al., 2024) due to variation in source and time of straw application and soil type, climate condition. Soil bacteria and fungi diversity were significantly increased by SBF, partially in line with Dong et al. (2017). The increase in soil bacterial and fungal diversity under the SBF may be attributed to the enhanced heterogeneity of the soil micro-environment induced by plastic film mulching. Variations in temperature, moisture, and oxygen availability across different soil layers create distinct niches, which can support the proliferation of microorganisms occupying diverse ecological roles (Li et al., 2024).

In our study, SBFSF was effective in enhancing numbers of bacterial genera with biocontrol potential and nutrient cycling properties. The SBFSF enriched Bacillus and Lysobacter, which correlated with the soil carbon and nitrogen cycling. Bacillus are also well known and have been reported to possess a potential for antagonism against soil-borne pathogens, mediated by the production of antimicrobial compounds (Saxena et al., 2020), this in turn could enhance crop productivity and carbon sequestration (Wang et al., 2015). The predicted functional linkage to the metabolism of other amino acids strengthens the inference for their greater contribution to N cycling and the biosynthesis of bioactive compounds. Similarly, Lysobacter can also lyse fungi and are involved in the decomposition of resistant organic matter (Afoshin et al., 2020). Their predicted functional profiles, associated with amino acid metabolism, lipid metabolism, and metabolism of other amino acids, suggested their multifunctional potential in C and N turnover. These findings suggested that SBFSF build a disease-suppressive soil environment with continuous nutrient cycling. In contrast, SM significantly increased Actinobacteriota and Chloroflexi but also enriched potentially pathogenic fungi. It is well known that members of Actinobacteriota can degrade complex lignocellulose and produce antibiotic compounds, which suppress the growth and reproduction of the pathogenic microorganisms (Yang et al., 2024; Ma et al., 2018; Zheng et al., 2022), Chloroflexi involving organic substance decomposition, carbon cycling to increase soil conditions making it more suitable for plant growth (Hu et al., 2023). At the genus level, SM promoted the substantial growth of JG30-KF-CM45, suggesting that JG30-KF-CM45 may play an important role in soil carbon and nitrogen cycling. However, the substantial increase in Fusarium, Cladosporium, and Alternaria - species all containing known pathogens causing various disorders of tomato like wilt, leaf spot and fruit rot (Zhao L. et al., 2023; Kong et al., 2023; Zhang et al., 2025)—were alarming with respect to the possible risk potential of diseases due to SM. Thus, while the effect of SM on beneficial decomposers such as Actinobacteriota and Chloroflexi are promoted, the increase in plant potential pathogens appears to counterbalance those beneficial ecological effects. SBF significantly enriched Acidobacteria and Vicinamibacteraceae, taxa commonly associated with acidic environments (Gonçalves et al., 2024), a finding consistent with the decreased soil pH observed under SBF in our study. Furthermore, SBF increased the relative abundance of Fusarium and Cephaliophora. Of particular note, a strain of Cephaliophora known to cause tomato leaf spot was first reported in the United States in 2023 (Luo et al., 2020), suggesting that long-term silver-black plastic film mulching may also promote the proliferation of potential pathogenic fungi. It is noteworthy that the identification of the aforementioned pathogens is based on an analysis of relevant literature, and their precise functions require further experimental validation.

Fungal community analysis revealed that SBFSF increased the abundance of Ascomycota, Mortierellomycota at the phylum level and Trichocladium at the genus level, which are involved in decomposition of cellulose and chitin (Li et al., 2020; Ozimek and Hanaka, 2021; Wu et al., 2022). More importantly, functional predictions indicated that SBFSF enhanced multiple metabolic pathways including electron transfer, respiration, vitamin biosynthesis, and secondary metabolite production. These improvements indicate a more metabolically active fungal community, which is capable to sustain soil ecosystem functioning.

The differing responses of microbial community to mulching practices indicates a balance between diversity and function maintenance. Even if SM and SBF promoted taxonomic diversity, SBFSF generated a functionally specialized community with greater metabolic potential and lower potential pathogen load. This is indicative that the functional attributes may be more important than overall diversity in supporting soil health and plant productivity.

Mulching is widely utilized in field crop and vegetable production and has been shown to promote crop growth (Sharma et al., 2023; Wang J. et al., 2025; Zhao Y. et al., 2023). Similarly, our study demonstrated that all mulching treatments significantly enhanced tomato yield compared to CK, with SBFSF exhibiting the most pronounced effect. These improvements can be attributed to the integrated modifications in plant growth and soil microenvironment induced by the different mulching practices.

The higher yield under SBFSF compared to SBF and SM was also consistent with enhanced dry matter accumulation and root growth. Increased root length and area under SBFSF probably contributed to enhanced nutrient and water uptake (Mooney et al., 2010; Guo et al., 2024) which resulted in higher biomass production and fruit yield. The early advantage in dry matter under SBF at 20 days may be attributed to its more effective soil warming during initial growth. However, SBFSF surpassed all other treatments by 40 days, indicating that its combined regulation of temperature and moisture better sustained plant development. Furthermore, SBFSF promoted the proliferation of beneficial soil microorganisms such as Firmicutes, Bacillus, Lysobacte, Ascomycota, Mortierellomycota and Trichocladium, consequently, the beneficial microorganisms improved key soil properties, including soil organic matter content, total nitrogen, and available nutrients, ultimately leading to an increasing in tomato production (Tao et al., 2024). These findings align with previous reports demonstrating that microbial community modulation can enhance crop productivity (Wang et al., 2024; Li et al., 2020). Conversely, the stimulation of potential pathogenic species (Fusarium, Alternaria, Cladosporium) under SM may have partially counteracted the benefits of nutrient availability, potentially explaining why tomato production under SM underperformed compared to SBFSF although they had improved some soil properties. Similarly, the SBF treatment also elevated the abundance of potentially pathogenic fungi, including Cephaliophora and Fusarium. Given its suboptimal soil physicochemical performance, this likely contributed to the reduced yield of SBF compared to SBFSF. Based on partial least squares path modeling (PLS-PM), it suggested that soil properties were strongly and directly associated with tomato growth, whereas mulching practices were linked to enhanced tomato growth primarily through their association with changes in bacterial and fungal diversity and community structure, which in turn were related to soil properties.

Our study revealed an increase in the population of soil potential pathogenic fungi following 3 years of plastic film mulching alone and straw mulching alone. Despite this increase, tomato yields remained higher compared to the control. Thus, the potential continuous decrease in tomato yield over successive planting years warrants further observation over an extended period. Our study spanned a limited duration of 3 years; moving forward, our research will persist in examining both tomato soil quality and yield. Additionally, we examined soil nutrient and microbial conditions only before tomato vine removal and did not track their dynamics across multiple time points. Investigating those temporal changes is a direction for our future research.