Introduction

Accompanied by the constant increase in crop demand, the annual production of crop straw reached approximately 4 billion metric tons globally1,2. Crop straw is widely used for soil amendment, animal feeds, cooking and heating, and feedstock for biofuels3,4. Along with these multipurpose uses, straw incorporation is widely recommended as environmentally friendly management for sustainable agriculture, to ensure sustainable crop production and climate change mitigation5,6,7. In the coming years, the projected 9.8 billion people and the doubling of food demand require a holistic and comprehensive solution8,9. Inappropriate agricultural management (i.e. over-tillage and over-use of fertilizer) to achieve high crop yields, more than 30% of the world's agricultural land experiences land degradation10. Therefore, sustainable soil management strategies are needed to feed the increasing world population without exacerbating the environment11. Integrated investigations of straw incorporation's impact on soil physicochemical and biological properties, and crop performance may enhance our understanding of sustainable agriculture.

Straw return is widely adopted as a field management technique for sustainable agriculture. It has numerous advantages related to ecosystem services, including enhancing the soil nutrient supply and ensuring sustainable crop productivity4,9,12. According to a global-scale meta-analysis, straw return led to an increase in soil organic carbon (SOC) of more than 12%13. Following straw return in Italian rice cropping systems, there was an increase in soil total nitrogen (TN) and nitrogen utilization efficiency9. The decomposition of crop residues enhances the availability of nutrients essential for crop growth9. In a meta-analysis, Xia et al.14 discovered that straw return to global agroecosystems resulted in a 14% increase in available nitrogen, a 10% increase in phosphorus, and an 18% increase in potassium levels in the soil. Furthermore, returning straw to the fields can enhance soil structure by reducing bulk density (BD), increasing porosity, and regulating soil temperature and humidity, consequently improving water use efficiency and grain yield2.

Extracellular enzymes in the soil are primarily synthesized and secreted by soil microorganisms and serve as crucial facilitators of organic matter decomposition and nutrient cycling12,14. Zhang et al.4 reported that soil microbial activity was increased with straw incorporation, resulting in higher activities of soil urease, invertase, and phosphatase. Zhao et al.9 showed that the activities of β-1,4-glucosidase, β-1,4-xylosidase and N-acetylaminoglucosidase were increased by 11–21%, 19–33%, and 22–33%, respectively, in calcareous soils by returning straw to the soil. Furthermore, in comparison to fields without straw return, straw return resulted in notable increases in soil organic matter fractions, such as total and particulate organic C, N, and soluble organic C, as well as increased enzyme activity15.

Inner Mongolia is an important grain production base in the country and plays an important role in guaranteeing a sustainable increase in corn production capacity and food security. Tumochuan Plain Irrigation District (TPID), belongs to the temperate continental monsoon climate, its soil type is mainly sandy loam, is one of the main corn-producing areas in Inner Mongolia16. It is a typical agricultural area dominated by continuous maize cultivation, and straw return to the field is common. It has been shown that under long-term no-tillage conditions, the content of macroaggregates in the soil increases significantly, which in turn increases soil compactness17,18. Soil acidification subsequently occurs as a consequence of the stratification of SOC and nutrients, consequently impacting crop growth17,19. Zhang et al. revealed that the average annual yields of wheat and rice in a no-tillage straw mulch return system (14 t ha−1) were lower than those in a no-return plowed system (15 t ha−1) and a return plowed system (16 t ha−1) in sandy loam soils in Jiangsu Province17. A higher soil capacity caused by no-tillage limits rice yield20.

Crop yields exhibit varying responses to tillage and straw management. These diverse responses may be correlated with disparities in soil types, cropping systems, climatic conditions, and land use practices, all of which can impact soil nutrient distribution, fertility levels, and physical structure. Thus, the effects of different straw return methods on grain yield and soil fertility are highly variable and strongly dependent on site-specific conditions and sound agronomic practices21. Therefore, this study aimed to (1) determine whether different straw return methods can improve soil structure and fertility (i.e., soil organic carbon, total nitrogen, and quick-acting nutrients) and reveal the enzyme-mediated mechanism of nutrient release from straw. (2) Investigate the effects of different straw return methods on crop yields. (3) Clarify the relationships between soil quality and crop yields under different straw return methods.

Materials and methods

Experimental site

This study is based on the pre-constructed straw return trial platform of the group (to be implemented in 2018) starting in 2021–2022 and was carried out in the China Cilechuan Modern Agricultural Expo Park (Beizhitu Village, Goumen Town, Tumet Right Banner, Baotou City, Inner Mongolia, latitude 40°28′28″ N, longitude 110°29′5″ E), where perennial straw return to the experimental field was implemented starting in 2018. This area has a semiarid mesothermal temperate continental monsoon climate, with an average annual temperature of 6–8 °C, 400 mm of annual precipitation, a frost-free period of 140 days, an elevation of 1015 m, an annual number of sunshine hours of 2806 h, and an annual active cumulative temperature of 3000–3500 °C. The site is used for continuous maize cultivation. In the absence of tillage, the soil texture is sandy loam, and the soil fertility is characterized by an organic matter content of 22.04 g/kg, an alkali-hydrolyzable nitrogen content of 57.82 mg/kg, an available phosphorus content of 3.57 mg/kg, and an available potassium content of 84.97 mg/kg. The soil nutrient data collected before sowing and tillage (0–45 cm soil layer) are shown in Table 1, and the main meteorological data collected during the test period are shown in Fig. 1.

Table 1 Soil nutrients under different soil treatments in 2021.
Full size table
Figure 1
figure 1

Main meteorological factors during the growing period in the experimental area.

Full size image

Experimental design

The experiment used a one-way experimental design, where the plowing method was applied in the central zone. The shallow rotation (CK) method was used to control the farmers, referring to the sowing methods of local farmers, and sowing was performed with conventional shallow rotation in spring after straw stubbing. Three treatments were established for comparison, straw incorporated with deep tillage (DPR), the straw is crushed twice in full quantity when the land is prepared in autumn, and the soil is turned over 30–40 cm deep by deep tilling plow. Turn the soil 30–40 cm, plow the straw into the soil, and sow the seeds with a conventional planter in the following spring; straw incorporated with subsoiling (SSR), when preparing the land in the autumn, the whole amount of straw was crushed for the second time, and the soil was deeply loosened for 35–40 cm using a deep loosening machine. The straw was mixed with the soil, and the seed was sown with a conventional planter in the spring of the following year; no-tillage mulching straw return (NTR), when preparing the land in the autumn, the whole amount of straw was crushed for the second time and covered on the surface of the soil, and the seed was sown with a no-tillage planter in the spring of the following year. All straw return treatments included full corn straw return at 135 ~ 150 kg ha−1. The maize variety planted was Xianyu 696 at a planting density of 825 plants/ha. There was a total of four treatments with three replications per treatment. Hence. The experiment consisted of 12 plots, each plot measuring 65 m × 6 m. Fertilizer was applied at 3 kg ha−1 pure N, 1.5 kg ha−1 P2O5 and 1.5 kg ha−1 K2O. N was applied at a ratio of 3:7 at the nodulation and maceration stages, and P2O5 and K2O were applied once as basal fertilizers. The other management practices used were the same as those used in the study region.

Soil sample collection

The experiment was conducted in May 2021 and May 2022 during the pre-planting period of maize (the fourth and fifth consecutive years of implementation) using the "S" sampling method. Soil samples were collected from the 0 to 15 cm, 15 to 30 cm, and 30 to 45 cm soil horizons in each plot using a soil auger (TG-25; Shaoxing Bowei Equipment Co., Ltd., China) and sieved through a 2 mm sieve to remove plant and animal residues and other debris. The samples were subsequently sieved through a 2 mm sieve to remove plant residues, animal residues, and other debris. Then, the excess remaining soil samples were removed, collected by the quadratic method, put into a sealed bag, and immediately transported to the laboratory. Soil samples were divided into three portions, one of which was air-dried for the determination of soil physicochemical parameters and the activities of soil alkaline phosphatase (ALP), leucine aminopeptidase (LAP), β-glucosidase (β-GC) and cattails (H2O2). One was immediately stored in a refrigerator at 4 °C for the determination of soil microbial biomass carbon (MBC), soil microbial biomass nitrogen (MBN), nitrate reductase (NR), glutamine synthetase (GS), and pyrophosphorylase (PPi) activities. The third was immediately stored at − 20 °C for the determination of soil ammonia monooxygenase (AMO) and ribulose-1,5-bisphosphate carboxylase (RUBP ) activities.

Indicators and measurement methods

Determination of soil physical properties

Soil bulk density (g/cm3) (BD) was measured by the core method22. Soil temperature (ST): measured with a JL-01 multi-point soil temperature and humidity recorder (JL-01, Jingyi Electronic Company). Soil total porosity (TP): was calculated from bulk density as described by Lin et al.23.

Measurement of soil chemical properties

Soil alkali-hydrolyzable nitrogen (AN) content was determined by the alkali hydrolysis diffusion method24. Soil available P (AP) concentration was determined by means of a Smartchem450 automatic chemical analyzer (Smartchem450, AMS France). Soil available potassium (AK) concentration was determined by a flame photometer (M410, SHERWOOD SCIENTIFIC COMPANY, USA)25. Soil pH was measured in a 1:2.5 (w/v) soil: water mixture using a pH meter (STARTER 2100/3C pro-B, OHAUS Corporation, USA). Soil organic carbon (SOC) was analyzed with H2SO4-K2Cr2O7 solution26. Total nitrogen (TN) was analyzed based on the Kjeldahl digestion method27. Soil nitrate nitrogen and ammonium nitrogen (NO3−–N and NH4+–N) were determined by a Smartchem140 automatic chemical analyzer (SMARTCHEM140, AMS, France).

Measurement of the soil microbiota

Soil microbial biomass carbon (MBC) and nitrogen (MBN) were measured by chloroform fumigation–extraction based on the methods of Mori, with conversion factors of 0.45 and 0.54 for MBC and MBN, respectively28.

Measurement of soil enzyme activities

The activities of nine enzymes, including nitrogen cycle-related (N-acq) enzymes such as NR, AMO, GS, and LAP; phosphorus cycle-related (P-acq) enzymes such as ALP and PPi; and carbon cycle-related (C-acq) enzymes such as H2O2, RUBP, and β-GC, were determined. The colorimetric method (Bio-Rad iMark, Bio-Rad, USA) was used to assess C-acq enzymes and catalase activity was determined via potassium permanganate titration. In addition, the activities of enzymes within the same functional class were normalized. For instance, C-acq enzyme activity was assessed using the following Equation29:

$${\text{C}} - {\text{acq = }}\sqrt[3]{{H_{2} O_{2} *RUBP*\beta - GC}}$$
(1)

where H2O2, RUBP, and β-GC represent catalase, ribulose-1,5-bisphosphate carboxylase, and β-1,4-glucosidase, respectively.

Calculation of the soil quality index

The soil quality index (SQI) was estimated using the total dataset method for SQI evaluation30. First, all the soil properties were transformed into values from 0 to 1. A linear scoring model was used, described as follows:

$$\begin{array}{*{20}l} {{\text{Ascending}}\;{\text{ affiliation}}\;{\text{ function}}} \hfill & {{\text{XL}}\left\{ {\begin{array}{*{20}l} {0.1} \hfill & {{\text{X}} \le {\text{X}}1} \hfill \\ {0.9*({\text{X}} - {\text{X}}1)/({\text{X2}} - {\text{X}}1)*0.1} \hfill & {{\text{X1}} < {\text{X}}} \hfill \\ {1.0} \hfill & {{\text{X}} \ge {\text{X2}}} \hfill \\ \end{array} } \right.} \hfill \\ {{\text{Descending}}\;{\text{ affiliation}}\;{\text{ function}}} \hfill & {{\text{XL}}\left\{ {\begin{array}{*{20}l} {0.1} \hfill & {{\text{X}} \ge {\text{X2}}} \hfill \\ {0.9*({\text{X2}} - {\text{X}})/({\text{X2}} - {\text{X}}1)*0.1} \hfill & {{\text{X1}} < {\text{X}}} \hfill \\ {1.0} \hfill & {{\text{X}} \le {\text{X}}1} \hfill \\ \end{array} } \right.} \hfill \\ \end{array}$$
(2)

Where XL is the linear score (0 ~ 1), x is the measured value of the indicator, and × 2 and × 1 are the minimum and maximum values of the indicator, respectively. In particular, the pH and BD in all the treatments exceeded the optimal values (pH = 7.88, BD = 1.62 g/cm3) and were normalized by Eq. (2).

Next, the weight (Wi) of each indicator, i.e., the ratio of its variance to the cumulative variance, was determined via principal component analysis. The SQI was subsequently calculated using the following equation:

$$SQI = \sum\limits_{i = 1}^{n} {w_{i} \times S_{L} }$$
(3)

where Wi is the weight of the earth evaluation indicator, SL is the indicator score, and n is the number of indicators.

Grain yield

The measurement area per plot was 36 m2. Within each area, the number of harvested spikes was counted, and the fresh grain weight and water content were measured after manual threshing. The measurements were then converted to the yield at 14% water content. Additionally, ten spikes were left in each plot and air-dried. The number of grains in the spikes and the 100-grain weight, measured at 14% water content, were determined.

Statistical analyses

All the statistical analyses were performed using Microsoft Excel 2016. SPSS 25 (SPSS, Inc., Chicago, USA) statistical software was used to test the effect of a one-way ANOVA to investigate the effect of different straw return methods on soil quality and maize yield. Significance was tested using Tukey’s HSD (Honestly Significant Difference) at a 5% level of significance. Principal components analysis (PCA) was performed (‘Vegan’ package), and significant differences in enzymatic activities were tested using ADONIS (‘Vegan’ package) based on pairwise Bray–Curtis and weighted Unifrac dissimilarities. In addition, RDA was performed using Canoco5 (Cabit Information Technology Co., Ltd., Shanghai, China). Origin2022 (Northampton, UK) was used for plotting.

A structural equation model (SEM) was constructed to explore the direct and indirect impacts of soil properties on maize yield. This analysis was based on a multivariate approach using AMOS software (IBM SPSS AMOS 22.0). Predictors were obtained based on a correlation analysis of all measured indicators in 2021 and 2022. Model fitting was assessed using the chi-square method and the associated p value, comparative fitting index (CFI), goodness of fit (GFI), and root mean square error of approximation (RMSEA).

Ethical approval

The collection of plant material and the performance of experimental research on such plants complied with the national guidelines of China.

Results and analyses

Soil physicochemical properties

Impact of various straw return methods on soil physical properties

To determine the influence of tillage and straw return practices on soil and crop indicators, we first assessed the physicochemical properties of the soils in the DPR, SSR, NTR, and CK treatments. The soil moisture content (SM), porosity (total porosity, TP), temperature (ST), and bulk density (BD) exhibited significant variations among the different years, treatments, and soil horizons (Fig. 2A–D). Moreover, the soil BD exhibited highly significant differences among the treatments, soil horizons, and their interactions (p < 0.05).

Figure 2
figure 2

Effect of straw return methods on soil physical properties in the early stage of corn growth. DPR, straw incorporation with deep tillage; SSR, straw incorporation with subsoiling NTR, no-tillage mulching straw return; CK, farmer’s shallow rotation. Y, year; T, straw management; L, layer. Lowercase and uppercase letters indicate significant differences between treatments in 2021 and 2022, respectively (p < 0.05), as determined by the HSD test. The values are the means ± standard errors (n = 3).

Full size image

Soil bulk density (BD) was significantly reduced by 9.85% and 8.4% in the 0–45 cm soil layer in the straw deep plowing and returning methods (DPR and SSR) compared to the CK treatment, respectively, and by 8% and 6.9% compared to the NTR treatment, respectively, from 2021 to 2022. Soil SM and ST were significantly increased by 9.62–15.17% and 11.14–23.65% in 0–45 cm soil layer straw return (DPR, SSR, and NTR) treatments, respectively, compared to CK treatment. Soil TP was significantly increased by 14.21% and 11.47% in the 0–45 cm soil layer in the straw deep plowing method compared to the CK treatment, and by 11.70% and 9.02% compared to the NTR treatment.

Effects of different straw return methods on soil chemical properties

The soil available phosphorus (AP), available potassium (AK), organic carbon (SOC), ammonium nitrogen (NH4+–N), nitrate nitrogen (NO3−–N), alkali-hydrolyzable nitrogen (AN), and total nitrogen (TN) exhibited significant variations across different years and soil horizons (Fig. 3B–K). Additionally, the soil pH exhibited highly significant differences among the treatments and soil horizons (Fig. 3A) (p < 0.05).

Figure 3
figure 3

Effect of straw return on soil chemical properties in the early stage of corn growth. DPR, straw incorporation with deep tillage; SSR, straw incorporation with subsoiling NTR, no-tillage mulching straw return; CK, farmer’s shallow rotation. Y, year; T, straw management; L, layer. Lowercase and uppercase letters indicate significant differences between treatments in 2021 and 2022, respectively (p < 0.05), as determined by the HSD test. The values are the means ± standard errors (n = 3).

Full size image

Soil pH was not significantly different between treatments in 2021, and was significantly reduced by 10.28% and 18.50% in the 0–30 cm soil layer in 2022 for the DPR and SSR treatments, respectively, compared to the CK treatment. Soil AK, AP, SOC, AN, TN, NH4+–N, and NO3−–N straw-returned treatments were higher than CK treatments in 2021–2022, with soil AK, AP, AN and TN being the most prominent in the 0–15 cm soil layer (Fig. 3B–C,E–F); Soil SOC, NH4+–N and NO3−–N increased most significantly in the 30–45 m soil layer (Fig. 3D,G–K). Significantly 51.72–81%, 39.68–90.26%, 9.77–10.90%, 33–45.70%, 19.38–24%, 83.99–117.63%, and 20.71–30.17% were observed in the 0–45 cm soil layer in the straw-returned treatments as compared to the CK treatment, respectively (p < 0.05).

Effects of different straw return methods on soil microbial biomass, carbon, and nitrogen

As shown in Fig. 4A,B, there were highly significant differences in soil microbial biomass carbon (MBC) and nitrogen (MBN) among the years, treatments, and soil horizons from 2021 to 2022 (p < 0.05). Soil MBC in the 0–45 cm soil layer was significantly higher in the straw return treatments compared to CK treatment by 54.71–68.57% and 15.80–26.44% in 2021 and 2022, respectively; Soil microbial biomass nitrogen was significantly higher in the 0–30 cm soil layer of straw-returned treatments (DPR, SSR, and NTR) compared to CK treatment by 8.78–19.36% and 19.78–29.64% in 2021 and 2022, respectively (p < 0.05).

Figure 4
figure 4

Effects of straw return mode of soil microbial biomass carbon (MBC) (A) and nitrogen (MBN) (B) in the early stage of maize sowing. DPR, straw incorporation with deep tillage; SSR, straw incorporation with subsoiling NTR, no-tillage mulching straw return; CK, farmer’s shallow rotation. NTR, mulching straw return to the field; SSR, deep mixing straw return to the field; DPR, deep plowing straw return to the field; CK, conventional shallow rotation. Y, year; T, straw management; L, layer. Lowercase and uppercase letters indicate significant differences between treatments in 2021 and 2022, respectively (p < 0.05), as determined by the HSD test. The values are the means ± standard errors (n = 3).

Full size image

Influence of various straw return methods on soil enzyme activities

Dynamic changes in soil enzyme activity characteristics under various straw return methods

Variations in soil enzyme activities were observed among the treatments (Fig. 5A). Highly significant differences were found in the enzymes associated with soil carbon cycle (C-acq), phosphorus cycle (P-acq), and nitrogen cycle (N-acq) across years, tillage practices, soil horizons, and the interaction between tillage practices and interannual variability (p < 0.05).

Figure 5
figure 5

Heatmaps of various enzyme activities under different straw return methods. The circles represent the coefficient of variation (CV) of enzyme activity as an indicator of sensitivity to straw return strategies. Color ranging from blue to red indicates enzyme activity from low to high. DPR, straw incorporation with deep tillage; SSR, straw incorporation with subsoiling NTR, no-tillage mulching straw return; CK, farmer’s shallow rotation. Y, year; T, straw management; L, layer. Lowercase and uppercase letters indicate significant differences between treatments in 2021 and 2022, respectively (p < 0.05), as determined by the HSD test. The values are the means ± standard errors (n = 3).

Full size image

The levels of soil C-acq in the straw return treatment were greatest in the 0–15 cm soil layer (Fig. 5B). In the 0–45 cm soil layer, compared with those in the CK treatment, the soil C-acq in the DPR, SSR, and NTR treatments significantly increased by 15.15–42.65% and 35.99–95.39% in 2021 and 2022, respectively (p < 0.05).

The soil N-acq and P-acq in the straw return treatments were greatest in the 30–45 cm soil layer (Fig. 5C,D). In the 0–45 cm soil layer, the soil N-acq and C-acq levels in the straw deep plowing and returning methods were significantly greater than those in the CK treatment, with increases of 121.7–190.6% for N-acq and 94.37–102.86% for P-acq in 2021 and 130.05–216.77% for N-acq and 139.30–147.96% for P-acq in 2021 (p < 0.05).

Effect of soil physicochemical properties on soil enzyme activities

Principal component analysis (PCA) indicated that in the 30–45 cm soil layer, the enzyme activity in the CK treatment differed from that in the straw return treatments and was primarily associated with PC1 (Fig. 6E). PC1 and PC2 explained 73.32% and 12.03%, respectively, of the variance. In the 0–15 cm soil layer, nitrate nitrogen was linked to soil enzyme activities and accounted for 63.3% of the total enzyme variation (Fig. 6A,B). In the 15–30 cm soil layer, the MBC was linked to soil enzyme activity and accounted for 55.6% of the total enzyme variation (Fig. 6C,D). In the 30–45 cm soil layer, the MBC was associated with soil enzyme activity, explaining 68.4% of the total enzyme variation (Fig. 6F).

Figure 6
figure 6

Principal component analysis of soil enzyme activity in the three soil layers under the different straw return modes. Redundancy analysis (RDA) of soil enzyme activity and soil physicochemical properties in the three soil layers under the different straw return modes. DPR, straw incorporation with deep tillage; SSR, straw incorporation with subsoiling NTR, no-tillage mulching straw return; CK, farmer’s shallow rotation. The black arrows represent soil physicochemical properties, and the red arrows represent soil enzyme activities.

Full size image

Soil quality index and maize yield

Figure 7 shows that the soil quality index (SQI) exhibited highly significant differences across years, tillage practices, and their interaction (p < 0.05). The SQI of the straw return treatments (DPR, SSR, and NTR) exceeded those of the CK treatment by 2.41, 2.17, and 1.61 units in 2021 and by 3.56, 3.02, and 2.33 units in 2022, respectively.

Figure 7
figure 7

Soil quality indices (SQIs) under the different straw return strategies. DPR, straw incorporation with deep tillage; SSR, straw incorporation with subsoiling NTR, no-tillage mulching straw return; CK, farmer’s shallow rotation. Y, year; T, straw management; L, layer. Lowercase and uppercase letters indicate significant differences between treatments in 2021 and 2022, respectively (p < 0.05), as determined by the HSD test. The values are the means ± standard errors (n = 3).

Full size image

As illustrated in Fig. 8A, maize yield exhibited highly significant differences across years, tillage practices, and their interaction (p < 0.05). In 2021, the maize yield was significantly greater in the DPR, SSR, and NTR treatments than in the CK treatment by 9.94%, 8.82%, and 3.66%, respectively, and by 13.14%, 11.41%, and 5.96%, respectively, in 2022. The maize yields in both 2021 and 2022 were highest in DPR (16.41 t ha−1), followed by SSR (16.20 t ha−1), NTR (15.42 t ha−1) and CK (14.72 t ha−1). More importantly, a significant positive correlation was observed between the SQI and maize yield (R2 = 0.86, p < 0.01; Fig. 8B).

Figure 8
figure 8

Maize yields (A) and the relationship between the soil quality index (SQI) and average maize yield (B). In both 2021 and 2022. Lowercase and uppercase letters indicate significant differences between treatments in 2021 and 2022; DPR, straw incorporation with deep tillage; mulching straw return to the field (red box); SSR, straw incorporation with subsoiling (green box); NTR, no-tillage mulching straw return (blue box); CK, farmer’s shallow rotation (yellow box). Y, year; T, straw management; L, layer. The white squares represent the grain output in the past two years. Lowercase letters indicate significant differences between the different straw return methods at the 0.05 level. The values are the average values ± standard errors (n = 3).

Full size image

Quantitative contribution of soil characteristics to maize yield

Structural equation modeling (SEM) methods were used to analyze the direct and indirect effects of straw return methods on soil physical properties, soil chemical properties, soil microbial carbon, and soil enzyme activities on maize yield (Fig. 9). The results demonstrated the strong positive direct effect of the straw return method on the MBC (0.44); soil enzyme activity, including that of RUBP, LAP, and PPi (0.592); and maize yield (0.431). Additionally, a strong negative direct effect on soil BD was observed (p < 0.001; − 1.372). Soil BD was strongly negatively correlated (p < 0.001) with SM, while MBC had a significant positive correlation (p < 0.05) with soil enzyme activity (0.427). SM (0.736), MBC (0.963), and enzyme activity (0.274) exhibited strong direct positive effects on soil nutrients. Additionally, soil nutrients (0.333) and straw return method (0.431) were positively correlated (p < 0.001) with crop yield, while soil BD (0.292) had a pronounced negative direct effect (p < 0.001) on maize yield. Overall, the SEM explained 97% of the total variation in maize yield. In conclusion, the straw return has dual effects. Initially, SM is enhanced by reducing soil BD, thereby increasing soil nutrients and impacting maize yield. Subsequently, through increased soil MBC, straw return triggers the release of RUBP, LAP, and PPi in the soil, accelerating the accumulation of nutrients and ultimately improving maize yield.

Figure 9
figure 9

Structural equation model (SEM) revealing the underlying mechanism of the relationship between straw return and maize yield. The solid and dashed arrows indicate positive and negative relationships, respectively. The number on each arrow indicates the normalized path factor, while the width of the arrow indicates the strength of the normalized path factor. R2 denotes the percentage of variance explained by the model. *, **, and *** represent p < 0.05, p < 0.01, and p < 0.001, respectively. RUBP, ribulose-1,5-bisphosphate carboxylase; LAP, leucine aminopeptidase; PPi, pyrophosphorylase; AK, soil available potassium; AP, soil available P; SOC, soil organic C; TN, total nitrogen; NO3−–N, soil nitrate nitrogen.

Full size image

Discussion

Effects of different straw return methods on soil quality

Soil physical conditions significantly impact crop production by influencing soil carbon levels, aeration, nutrient dynamics, and the penetration of crop roots. Nevertheless, the extended use of rotary or shallow plowing for straw return frequently results in subsoil compaction and reduced permeability, thereby obstructing the movement of water, air, and fertilizers between the tillage layer and the subsoil13,18. The current study revealed the following. This indicates that due to the synergistic impact of deep plowing and straw return, the combination of these methods serves as an effective soil management practice for enhancing soil structure31. Additionally, the results suggest that consecutive years of no-tillage straw return results in innate soil compaction and, consequently, an increase in soil bulk density32. In this study, it was found that DPR and SSR treatments significantly reduced soil BD and increased TP in the 15–45 cm soil layer compared to NTR (Fig. 2A). This suggests that the combined effect of deep plowing and straw return is an effective management practice for improving soil structure33. The straw return treatment significantly increased the water retention of soil in the 0–45 cm soil layer compared to CK. This phenomenon may be attributed to the fact that returned straw typically forms a barrier against evaporation at the soil surface and diminishes heat loss from the soil to the air, consequently preserving soil moisture and increasing the soil temperature32,34. The DPR treatment exhibited the highest soil temperature, followed by the SSR and NTR treatments, with the CK treatment exhibiting the lowest temperature. This may be because straw return usually creates a barrier against evaporation at the soil surface and reduces heat loss from the soil to the air, thus retaining soil moisture and increasing soil temperature35,36.

Returning straw to the field is usually considered an effective option for the enhancement of SOC pool and nutrient supply37,38. However, there were no differences in SOC, TN, and available nutrients among different straw management in the topsoil (0–20 cm) (Figs. 5 and 6). Hassink proposed that soil has a finite capacity to store organic matter for decades or more39. Nowadays, many researchers demonstrated that organic matter input is becoming less efficient as the soil is approaching its saturation capacity, which is in contrast with the previous concept that SOM storage was linearly related to exogenous organic matter input at steady state40,41. Generally, the distribution of nutrients in soils is not uniform with higher SOM content in the topsoil38. Long-term straw incorporation with rotary tillage in the NCP likely results in C saturation in the topsoil. Zhao et al. (2018a) suggested that single-crop straw incorporation was sufficient to maintain the initial SOC level in the maize-wheat rotation system, and the remaining straw could be used for other purposes42. Similarly, our results indicate that maize straw return is sufficient to maintain fertility (i.e., organic carbon, total nitrogen, and fast-acting nutrients) in the topsoil layer. Therefore, corn straws could be placed towards the lower soil layers. The deep plowing straw return treatment turned all the stalks uniformly into the 0–40 cm soil depth, and the organic carbon and total nitrogen contents in the 30–45 cm soil layer were significantly increased compared to the NTR treatment (Fig. 3). Generally, deep layers are far from the point of saturation, and thus have a greater SOC storage potential40. The MBC content was also significantly increased under deep plowing to return straw to the field in the subsoil (Fig. 4), indicating that the improved environmental conditions (e.g. soil moisture, temperature, air permeability) in the straw layer are suitable for microbial activities and biochemical reactions41. Thereby, straw decomposition was promoted, resulting in the accumulation of SOC and TN in 30–45 cm. Furthermore, the decreased BD and increased TP in subsoil under deep plowing under straw treatment might cause higher root growth and thereby increase the quality and quantity of organic matter input via root exudation and litter, further increasing crop yields43. Soil N transformation and crop uptake of N are highly dependent on agricultural management44. Deep plowing under straw treatment considerably increased soil available N (NO3−–N and NH4+–N) in the 30–45 cm soil layer (Fig. 3K–G), which was consistent with Yang et al. That the straw layer under ditch-buried straw incorporation was beneficial to the retention of available N45.

Enzyme-mediated nutrient release from straw

Soil extracellular enzymes play an important role in nutrient cycling and are strongly affected by straw return46. The SSR and DPR treatments significantly increased the activities of soil enzymes involved in C, N, and P cycling in the 0–45 cm soil layer compared to those in the CK treatment. Additionally, these treatments significantly boosted soil enzyme activities in the 15–45 cm soil layer compared to those in the NTR treatment (Fig. 5A–C). This difference was primarily attributed to the SSR and DPR treatments elevating soil microbial activity in the 15–45 cm layer, subsequently promoting increased synthesis and secretion of soil extracellular enzymes. Straw return is directly associated with MBC and MBN. This study revealed that, in comparison with the NTR treatment, the DPR and SSR treatments increased the carbon and nitrogen sources in the subsoil, providing nutrients for enhanced microbial growth and subsequent reproduction47. When straw is tilled into the 0–45 cm soil layer, additional nutrients and improved soil structure will provide energy and a suitable environment (e.g., soil temperature, moisture, and oxygen) for microbial growth and lead to an increase in the number of endoenzymes in the living microbial population48. Thus, more extracellular enzymes related to C, N and P were accumulated in 15–45 cm than in the NTR-treated deep-tilled straw return. Soil enzymes are proximate factors for SOC formation and decomposition and can break down polymers into monomers by reducing reaction activation energy49. There was a positive correlation between MBC and enzyme activity among different treatments (Fig. 6). SEM showed that enzyme activity directly and positively affected soil nutrients (Fig. 8). Thus, the deep ploughing straw return treatment promoted the accumulation of C-, N- and P-acquiring enzymes straw decomposition and nutrient release in the 30–45 cm straw layer compared to the NTR treatment, which led to an increase in soil nutrients (i.e., organic carbon, total nitrogen, and fast-acting nutrients) in the 30–45 cm soil layer.

Potential mechanisms underlying improved maize yield due to straw return

Compared with those in the CK treatment, the maize yield in each of the return treatments significantly (p < 0.05) increased, with the DPR and SSR treatments being the most prominent (Fig. 8A). These findings parallel the observations of Yin et al. who reported a notable increase in crop yield potential with straw return to the field, similar to the results of this study15. Numerous factors contribute to the mechanisms causing enhanced crop yields. The structural equation modeling analysis in this study revealed that soil BD and nutrients directly influence crop yield, accounting for 99.7% of the variation in crop productivity (Fig. 9). Initially, infiltration was promoted under the NTR treatment, consequently reducing runoff and soil surface evaporation while also maintaining or increasing soil organic matter. This promotes the maintenance of stable plant water requirements and a stable soil water supply during drought conditions15. Furthermore, the DPR and SSR treatments established favorable environments for microbial and enzyme activity, consequently enhancing the soil structure and moisture conditions, which significantly accelerated straw decomposition and increased the soil nutrient levels50. Additionally, the DPR and SSR treatments disrupted compacted subsoil and deepened the tillage layer, consequently enhancing the soil structure in the 0–45 cm soil layer and expanding the effective soil volume for root growth15,32. Moreover, burying maize straw in deeper soil layers can mitigate the adverse impacts of nitrogen competition between soil microorganisms and maize plants51. Given the degradation of soil health and the significant challenges associated with food security in China, the incorporation of straw back into fields has been suggested as an effective approach for enhancing soil fertility and boosting crop yields33. This study illustrates that DPR and SSR management strategies are viable for continuous maize cropping systems and improve crop productivity by enhancing physical, chemical, and biological processes in the subsoil. In addition, only three different straw management practices were compared in this study: straw incorporation with deep tillage, straw incorporation with subsoiling and no-tillage mulching straw return. This study examined only 2 years (2021 and 2022) after three years of straw return for a very short period of time. Further long-term investigations to compare the impacts of more straw management practices on agroecosystem productivity and soil quality (e.g., deep-injected straw and strip deep rotation straw return with straw incorporation combined with deep tillage incorporation) are essential for achieving a complete evaluation of the impacts of straw management practices on agroecosystem productivity.

Conclusion

The DPR and SSR treatments enhanced soil water retention by reducing the soil bulk density (BD) in the 0–45 cm soil layer. Similarly, compared with those in the NTR treatment, the soil organic carbon (SOC), total nitrogen (TN), readily available nutrients (NO3−–N), available potassium (AK), and available phosphorus (AP) in the 15–45 cm soil layer in the DPR and SSR treatments significantly increased. Additionally, no significant differences were observed among the straw return treatments within the 0–15 cm soil layer. This result was primarily attributed to the accumulation of MBC, as well as carbon, nitrogen, and phosphorus cycle-related enzymes, which facilitated straw decomposition and subsequent nutrient release after the straw was uniformly incorporated into the 0–45 cm soil layer under the DPR and SSR treatments. There was a strong positive correlation between the SQI and maize yield, indicating that enhancing soil quality through straw return contributed to increased maize yield. The maize yields in the DPR and SSR treatments were significantly greater than those in the NTR and CK treatments, particularly in the second year, which was attributed to the combined influence of soil ecological processes. Collectively, these findings indicate that the DPR and SSR strategies may serve as effective cultivation practices for enhancing soil quality and increasing maize yields in maize continuous cropping systems in irrigated areas of the Tumochuan River Plain.