Managing Residue Return Increases Soil Organic Carbon, Total Nitrogen in the Soil Aggregate, and the Grain Yield of Winter Wheat

Abstract

Soil tillage and maize residues return are important practices for tackling and promoting soil quality and improving crop yield in the North China Plain (NCP), where winter wheat production is threatened by soil deterioration. Although maize residues incorporation with rotary tillage (RS) or deep plowing tillage (DS) is widespread in this region, only few studies have focused on rotation tillage. Four practices, namely RT (continuous rotary tillage without maize residues return), RS, DS, and RS/DS (rotary tillage every year and deep plowing interval of 2 years), were evaluated under field conditions lasting a period of 5 years. After a 5-year field experiment, the mean soil bulk density of the 0–30 cm soil layer decreased significantly with RS, DS, and RS/DS, i.e., by 4.19%, 6.33%, and 6.71% compared with RT, respectively. The treatments greatly improved the total soil porosity, soil aggregate size distribution, soil aggregate stability, and the root length density in the 0–30 cm soil layers. Residues return with DS and RS/DS treatments significantly increased the soil organic carbon (SOC) and total nitrogen (TN) storage in the 0–30 cm soil layer, mainly owed to the increases in the SOC and TN pool associated with the macro-aggregate. A positive trend in the grain yield was noted under both DS and RS/DS conditions, whereas a decreasing tendency was presented in continuous rotary treatments. In summary, RS/DS treatment significantly increased the amount of SOC and TN, improved the particle size distribution of soil aggregates, and thus improved the soil’s physicochemical properties, which is beneficial for wheat to achieve high yields. Our results suggested that RS/DS was a highly efficient practice to improve soil quality and increase crop production in the NCP.

1. Introduction

The North China Plain (NCP) is an important wheat-growing area in China, accounting for about 60% of wheat production [1]. In recent years, wheat production has continuously improved in this region. However, the application of large quantities of chemical fertilizers is still the most common agriculture management practice contributing to wheat yield increases [2]. Long-term excessive use of fertilizers and a lack of adequate soil conservation measures have resulted in a series of environmental problems, including land desertification, water pollution, and excessive nitrogen emissions. Recently, soil deterioration has gained the attention of farmers and governments [3]. Therefore, determining agricultural methods for improving the soil quality and maintaining high crop production is urgently required [4,5,6]. Many studies have shown that the addition of crop residues was an efficient agricultural measure for improving soil quality and crop production [7,8,9].

Conservation agriculture, which includes crop residue incorporation, is the main strategy for combating soil degradation [10,11,12,13]. Residue return is often incorporated through soil cultivation during the agricultural production process. However, the improper combination of straw returning and other cultivation methods leads to a deterioration in soil structure and an uneven distribution of nutrients [14,15], thereby hindering the development of good root systems [16].

Deep plowing and rotary tillage are customary tillage practices in the NCP [17]. Deep tillage requires the use of costly agricultural equipment, making it more expensive than rotary tillage. On the other hand, farmers are increasingly favoring rotary tillage due to its simpler operation and reduced costs for preparing the land. However, prolonged rotary tillage often results in tight subsurface soils that are not favorable for crop production [18].

Soil aggregates can protect soil organic carbon (SOC), affect the soil tilth [19], and determine the soil’s nitrogen (N) reserve [20]. Compared to residue removal, maize residue return with plowing tillage improved aggregation and other properties in soil, as well as increasing crop yield [7]. Similar results were also observed under different tillage methods, such as residue mulching with no-till [21], and residue return with rotary tillage [22,23]. He et al. [24] indicated that the return of crop residues for a period of 17 years substantially enhanced the levels of SOC and total nitrogen (TN) in different aggregate fractions across three representative cropland soils in China. Consequently, the incorporation of crop residues into the soil not only contributes positively to soil quality but also plays a crucial role in the stabilization of soil aggregates, in addition to its impact on the content of SOC and TN within soil aggregate fractions.

The aforementioned research primarily concentrated on the impact of a single soil cultivation method—be it no-till, rotary tillage, or plowing—on the soil’s structural integrity, the levels of soil organic carbon (SOC) and total nitrogen (TN), as well as the yield of grain crops.

However, few studies have been conducted on the effects of mixed tillage patterns (e.g., two years rotary tillage followed by one year of deep tillage) on SOC and TN stocks in soil aggregates of different size fractions, the relationship between root distribution characteristics and soil physical traits, or wheat yields. In the wheat corn rotation system, we hypothesize that a mixed tillage method is more conducive to the accumulation of nutrients in the topsoil and the improvement of physical properties, such as the two-year rotary tillage and one-year deep tillage method. Based on this, we conducted this study. We conducted this research with the following objectives: (Ⅰ) changes in SOC and TN storage, (II) quantification of the distribution of SOC and TN in different aggregate fractions, and (III) grain yield of winter wheat.

2. Materials and Methods

2.1. Site Description

The field-based locational experiment was carried out from October, 2016 to June, 2021 at the Research Station of Jia Sixie Agronomy College, Shandong, China (35°66′ N, 116°77′ E), which experiences a temperate continental monsoon climate. The cumulative temperature exceeding 10 °C and the duration of sunlight each year were at least 4713 °C and 2403 h, respectively. On average, the annual precipitation during the wheat growing season from 1971 to 2016 amounted to 272.62 mm. The soil was a sandy loam with a pH of 7.92 and an EC of 0.23 mS cm⁻¹. Prior to the experiment, the plot was routinely rotated with winter wheat and summer maize for three years, so there was little spatial variation in the soil properties across the cropland and aboveground biomass from the field was removed after harvesting. The chemical properties of the soil in the 0–30 cm till layer were as follows: rapidly available phosphorus = 8.60 mg kg⁻¹ (Olsen method) [8]; rapidly available potassium = 57.50 mg kg⁻¹ (Dirks–Sheffer method) [25]; soil organic matter (SOM) = 12.30 g kg⁻¹ (wet oxidation–redox titration method) [26]; total nitrogen (TN) = 0.91 g kg⁻¹ (Kjeldahl method) [27]; NO₃^--N = 16.93 mg kg⁻¹ (continuous flow analyzer method), and NH₄⁺-N = 5.64 mg kg⁻¹ (continuous flow analyzer method) [28]. Soil physical properties and the SOC and TN storages of the initial condition (IC) before sowing are summarized in

Table 1. Soil bulk density (SBD), porosity, aggregate particle size, soil organic carbon (SOC), and total nitrogen (TN) storages of the tilth soil (0–30 cm) in the experimental site before treating.

Depths (cm)	SBD (g cm−3)	Porosity (%)	Aggregate Particle Size (%)		SOC/TN Storage (t ha−1)
			>0.25 mm	<0.25 mm	>0.25 mm	<0.25 mm
0–10	1.32	49.91	77.34	22.66	7.47/1.03	3.21/0.50
10–20	1.38	47.71	64.82	35.18	5.98/0.89	3.02/0.49
20–30	1.42	46.07	62.48	37.52	5.00/0.84	1.73/0.15

.

2.2. Experimental Design

The experiment was a completely randomized block design with three replications per plot. The area of each plot was 4 m × 15 m. The experimental treatments were combinations of tillage and total aboveground residues returned from the summer corn and the remnants of winter wheat were cleared from the field after harvest in the five years (RT; five consecutive years of rotary tillage operation with residues applied, RS; five consecutive years of deep plowing tillage with residues applied, DS; two years of rotary tillage followed by one year of deep tillage with residues applied, RS/DS). The “Jimai 22” winter wheat was cultivated at a planting density of 2.25 × 10⁶ plants ha⁻¹, featuring row spacing of 25 cm. The sowing dates were 12 October 2016, 11 October 2017, 14 October 2018, 15 October 2019, and 15 October 2020, respectively. During each wheat growing season, a pre-plant application was made which included phosphorus (P₂O₅) at a rate of 105 kg ha⁻¹, supplied as calcium superphosphate, and potassium (K₂O) at a rate of 75 kg ha⁻¹, in the form of potassium chloride. Additionally, half the required nitrogen (N) was applied at a rate of 112.5 kg ha⁻¹, using urea as the nitrogen source for the treatment.

Basal fertilizers were broadcasted before soil cultivation. The remaining nitrogen fertilizer was applied manually at the wheat jointing stage. One day after harvesting the summer corn, land preparation was performed. The operation procedures are shown in

Table 2. Operation procedures of different tillage practices.

Tillage	Operation Procedure
RT	Remove the corn straw out of the farmland, apply base fertilizer, then rotate it by rotary cultivator (model: IGQN-200K-QY, manufacturer: YTO Group Corporation, from Luoyang city, Henan Province, China) twice with a depth of 10–12 cm, seed with common seeder after forming the border check
RS	All the maize straw returned to the plot, apply base fertilizer, then rotate it with a rotary cultivator (model: IGQN-200K-QY, manufacturer: YTO Group Corporation, from Luoyang city, Henan Province, China) twice with a depth of 10–12 cm, seed with common seeder after forming the border check
DS	After crushing corn stalks and returning them to the field, apply base fertilizer, then deeply plow the soil to a 25 cm depth with a turnover plough (model: ILFQ330; manufacturer: Runlian S&T development Co., Ltd., from Ji’ning city, Shandong Province, China), and perform rotary tillage twice with a rotary cultivator (model: IGQN-200K-QY; manufacturer: YTO Group Corporation, from Luoyang city, Henan Province, China) with a depth of 10–12 cm, seed with common seeder after forming the border check
RS/DS	RS refers to the above in the first two seasons → DS refers to the above in the third season → RS refers to the above in the last two seasons

.

2.3. Sampling and Measurement

Soil samples were collected with 10 cm increments up to a depth of 30 cm, after the wheat harvest during the 2020–2021 growing season, using a 54 mm diameter soil column sampler (XDB0401, Beijing New Landmarker soil Equipment Co., Ltd., Beijing, China), and mixed to produce a composite sample. A quarter of the composite sample (approximately 400 g) underwent a sieving process, passing through a 7 mm mesh to eliminate any clumps, with any pebbles and clumps exceeding 7 mm in size being discarded.

We conducted root sampling at depths of 0–10, 10–20, and 20–30 cm in the anthesis stage in the 2020–2021 growing season, and we referenced the method of Dai et al. [29]. A trench was excavated to facilitate access to each experimental subplot. Within each unit, root samples were collected from a designated plot, which was home to a cluster of plants. The dimensions of each sampling plot were 0.75 m in length, measured at a right angle to the rows, spanning across three rows of plants and 0.40 m in width, measured along the rows. At the soil level, all the plants within the sampling plot were severed; afterward, any loose plant material on the soil surface at the sampling location was cleared away. The collected samples were then carefully placed into mesh bags made of string.

Roots from each bag were meticulously cleaned by removing the soil with tap water and rinsing the samples to eliminate any organic residue and other impurities. The roots were then neatly arranged and floated on a shallow layer of water within a clear tray measuring 0.20 m by 0.30 m. Using a high-resolution Epson Perfection V700 Photo (Seiko Epson Corp., Suwa City, Nagano Prefecture, Japan) to determine root traits, the measure of root length density, denoted in units of millimeters per cubic centimeter of soil (mm cm⁻³), was ascertained by the method of dividing the cumulative length of the roots by the volumetric extent of the soil core sample from which they were extracted.

The soil bulk weight was determined by the ring knife method [30] and a soil core was collected randomly from each plot. Soil porosity was derived from the equation relating to bulk density (BD) and particle density (PD, 2.65 g cm⁻³), as follows:

Porosity (%) = (1 − BD/PD) × 100

(1)

The distribution of soil aggregates was evaluated by taking a 50 g sample and placing it on a set of stacked sieves with mesh sizes of 5, 2, 1, 0.5, and 0.25 mm, fitted to a soil aggregate analyzer (TTF-100, Shandong Rhine Optoelectronics Technology Co., Ltd., Weifang city, Shandong Province, China). These sieves were submerged in water and subjected to a vertical movement of 4 cm, oscillating at a rate of 30 cycles every 60 s, for a total duration of 2 min. The resulting aggregate size classes were categorized as greater than 5 mm, between 5 and 2 mm, between 2 and 1 mm, between 1 and 0.5 mm, between 0.5 and 0.25 mm, and less than 0.25 mm. The proportions of each size class were determined after drying the soil on the sieves and measuring their weight [7]. For a detailed examination of the carbon and nitrogen contents within these aggregates, the dried soil from each fraction was passed through a 0.149 mm sieve for further analysis.

The distribution of aggregate sizes was utilized to ascertain the proportion of aggregates exceeding 0.25 mm in diameter, referred to as R_0.25 [31]. The mean weight diameter (MWD) and geometric mean diameter (GMD) were also calculated, and the formulas are as follows:

$$R_{0.25}={\displaystyle \frac{M_{r>0.25}}{M_T}}$$

(2)

$$MWD={\sum }_{i=1}^{n}XiWi$$

(3)

$$GMD=exp\left({\displaystyle \frac{\sum _{i=1}^{n}Wi\ln Xi}{\sum _{i=1}^{n}Wi}}\right)$$

(4)

R_0.25 is the volumetric proportion of soil particles with diameters greater than 0.25 mm. W_i is the mass of the aggregate within a specific size interval, which is calculated as a percentage of the total dry mass of the soil samples analyzed. n denotes the total number of sieves used in the classification process and Xi denotes the average diameter of the soil aggregate collected on each sieve.

The concentration of SOC in all samples was determined by wet oxidation–redox titration [26]. The organic carbon was converted to carbon dioxide by hexavalent chromium ions (Cr⁶⁺) in the dichromate solution. Excess Cr⁶⁺ was determined by ferrous sulfate titration. To correct for the potential underestimation of organic carbon due to incomplete oxidation and the presence of other reducing agents such as iron (Fe²⁺), a correction factor of 1.1 was applied. The total nitrogen (TN) content in the soil was ascertained utilizing an automated Kjeldahl distillation–titration unit (Foss, Sweden).

The C/N content stored within a particular soil stratum of depth d was determined using the following calculation:

C/N = C/N concentration × BD × d × 10

(5)

where C/N storage refers to the total SOC/TN (t ha⁻¹), C/N concentration refers to the C/N content in a specific depth (g kg⁻¹), and d is the depth of soil (cm).

The C/N stock calculation was conducted as follows [32]:

C/N stock = C/N concentration × BD × d × fraction weight

(6)

where C/N expresses the C/N in each fraction class (t ha⁻¹), the C/N concentration refers to the C/N in each fraction size (g kg⁻¹); BD is bulk density (g cm⁻³), d is the depth of soil (cm), and the fraction weight is the % weight of the fraction in the whole soil.

The evolution of C/N storage within a specific soil layer of depth d and across various aggregate size fractions was computed in the following manner:

Evolution of C/N storage = C_T/N_T − C_B/N_B

(7)

C_T/N_T expresses the C/N storages in different treatments and C_B/N_B refers to the C/N storages of the baseline soils.

Entire wheat plants within an area of 10 m² from each experimental plot were utilized to assess the yield at the wheat harvest.

2.4. Statistical Analysis

One-way ANOVA was performed to test the different treatment effects on the soil properties and grain yield. Correlation and regression analysis between the root length density and the soil physical properties were determined. Variance analysis was undertaken to identify the variation sources of the grain yield. All data were analyzed using SPSS 19.0, developed by IBM Corporation in Chicago, IL, USA. The LSD method was used for significance analysis detection. The significance level was p < 0.05. Additionally, the graphical representations of the data were crafted by SigmaPlot version 10.0.

3. Results

3.1. Soil Bulk Density and Porosity

The variation in SBD during the five-year experimental period is shown in Figure 1. Compared with RT, the treatments with RS, DS, and RS/DS significantly decreased the SBD, i.e., by 7.03%, 1.56%, and 4.69% at the 0–10 cm depth; 4.23%, 7.04%, and 6.34% at the 10–20 cm depth, and 1.30%, 10.39%, and 9.09% at the 20–30 cm depth, respectively, where the SBD increased with the depth of the soil layer. Compared with initial condition (IC), RS and RS/DS significantly reduced the soil bulk density of the 0–10 cm and 10–20 cm soil layers, while RT significantly increased the soil bulk density of the 20–30 cm soil layers.

This study found that in the 0–30 cm soil layers, the DS and RS/DS treatments increased the total porosity after 5 years, while the RT and RS treatments decreased total porosity in the 20–30 cm soil layers (Figure 1). Compared with RT, the total porosity of the 0–10 cm soil layer under the RS, DS, and RS/DS treatments increased by 6.94%, 1.69% and 4.24%, respectively. The depth of 10–20 cm was 5.31%, 8.14%, and 7.71%, respectively. The depth of 20–30 cm was 1.72%, 13.91% and 12.66%, respectively.

3.2. Soil Aggregate Size Distribution

3.2.1. Soil Aggregate Particle Size

Figure 2 shows that after 5 years of residue incorporation, the macro-aggregates (R_0.25) increased significantly (p < 0.05) in the 0–30 cm layers with RS, DS, and RS/DS compared with the levels of RT, i.e., by 20.45%, 15.42%, and 18.68% at the 0–10 cm depth; by 11.86%, 22.10%, and 19.86% at the 10–20 dept, and by 7.99%, 20.18%, and 16.90% at the 20–30 depth, respectively. Compared with IC, RS/DS significantly increased the R_0.25 in the soil layers of 0–10 cm, 10–20 cm, and 20–30 cm by 6.20%, 12.25%, and 8.18%, respectively. A significantly higher level of R_0.25 was found in the RS at a 0–10 cm depth, whereas it was significantly lower in the DS and RS/DS at 10–20 and 20–30 cm depths. Meanwhile, the R_0.25 decreased correspondingly with the increase in soil depth.

As showed in Figure 2, high R_0.25 in the treatments with residue return were mainly due to the obviously higher particle size distribution in the >5 mm, 5–2 mm, and 2–1 mm soil aggregate. Meanwhile, the particle size distribution of >5 mm, 5–2 mm, and 2–1 mm, and 0.5–1 mm in RT decreased with the soil layer depth.

3.2.2. MWD and GMD

The effects of straw returning on the soil aggregate WMD and GMD are shown in Figure 3. In the three soil depths, the MWD of the straw returning treatment was significantly higher than that of control treatment (p < 0.05). In the 0–10 cm soil layer, RS/DS > RS > DS > RT with the increase in treatment time; in the 10–20 cm layer, RS > DS > RS/DS > RT, and in the 20–30 cm layers, RS >RS/DS >DS > RT.

There were some differences in the results for the GMD and the values were shown to be RS > RS/DS > DS > RT. Figure 3 showed that after five years, the GMD increased significantly at 0–30 cm depths with RS, DS, and RS/DS compared with the levels of RT, i.e., by 17.91%, 10.45%, and 16.42% at the 0–10 soil depth; by 36.23%, 17.39%, and 13.04% at the 10–20 cm soil depth, and by 35.19%, 7.14%, and 16.67% at the 20–30 cm soil depth, respectively. Meanwhile, in the soil layer of 20–30 cm, the MWD and GMD of the straw returning treatment were significantly lower than those of initial condition (IC).

3.3. Root Length Density

The effects of residue return and tillage methods on the soil RLD were significant (p < 0.05) (Figure 4). The order of RLD in the 0–10 cm soil layer was RS > RS/DS > DS > RT, and in soil layers 10–20 and 20–30, DS > RS/DS > RS > RT.

The correlation analysis demonstrated that the RLD was positively and significantly correlated (p < 0.01) with the soil total porosity and the proportion of soil macro-aggregate (Figure 5). Meanwhile, the SBD and the proportion of soil micro-aggregate were significantly (p < 0.01) and negatively correlated with the RLD (Figure 5).

3.4. SOC and TN Storage in Aggregate

3.4.1. SOC Storage

As shown in

Table 3. The SOC in soil aggregate under different treatments.

Soil Depth (cm)	Treats	Total SOC (t ha−1)	C Distribution in Aggregate (t ha−1)
			>5 mm	5–2 mm	2–1 mm	1–0.5 mm	0.5–0.25 mm	<0.25 mm	>0.25 mm
0–10	IC	7.47 c	0.67 bc	1.39 a	1.51 b	2.57 a	1.32 b	3.21 ab	7.47 b
	RT	9.25 b	0.51 c	0.72 b	1.32 b	2.05 b	1.11 c	3.54 a	5.71 c
	RS	11.12 a	0.85 ab	1.40 a	1.92 a	2.38 ab	1.39 b	3.18 ab	7.94 ab
	DS	11.25 a	0.65 bc	1.44 a	1.89 a	2.74 a	1.28 b	3.24 ab	8.00 ab
	RS/DS	11.14 a	0.93 a	1.30 a	1.97 a	2.50 a	1.70 a	2.73 b	8.41 a
10–20	IC	5.98 c	0.52 cd	1.02 c	1.33 c	2.12 a	1.00 ab	3.02 ab	5.98 c
	RT	8.58 b	0.43 d	0.59 d	1.23 c	1.84 c	1.12 a	3.36 a	5.22 d
	RS	8.42 b	0.87 a	1.69 a	1.62 b	1.87 bc	0.86 b	1.51 c	6.91 ab
	DS	9.99 a	0.67 b	1.30 b	2.15 a	2.24 a	0.85 b	2.79 b	7.20 a
	RS/DS	9.74 a	0.63 bc	1.01 c	1.64 b	2.09 ab	1.11 a	3.25 ab	6.49 bc
20–30	IC	5.00 d	0.42 a	0.96 b	0.94 c	1.81 ab	0.87 c	1.73 a	5.00 cd
	RT	6.38 c	0.22 b	0.51 d	0.88 c	1.62 ab	1.48 b	1.67 a	4.71 d
	RS	6.60 bc	0.25 b	1.15 a	1.23 b	1.56 b	1.06 c	1.35 a	5.25 c
	DS	6.93 ab	0.42 a	0.60 c	1.62 a	1.84 a	1.84 a	0.60 b	6.33 a
	RS/DS	7.07 a	0.37 a	0.62 c	1.76 a	1.71 ab	1.46 b	1.15 ab	5.92 b

IC, initial condition; RT, five consecutive years of rotary tillage with residues removed; RS, five consecutive years of rotary tillage with residues applied; DS, five consecutive years of deep plowing tillage with residues applied, and RS/DS, DS after RS at a two-year interval. Values followed by different letters within a column are significantly different at p < 0.05.

, residue return and tillage methods significantly affected the total organic carbon storage and its distribution of different aggregate components. Overall, the residue treatment increased the total soil organic carbon storage from 0 to 30 cm (20.13%, 21.52%, and 20.34%, respectively); in the 10–20 cm soil layer, the values were −1.89%, 16.46% and 13.55%, respectively, and in the soil layers from 20 to 30 cm, they were 3.43%, 8.59% (p < 0.05), and 10.87%, respectively. The increase in the total SOC storage was primarily due to a significant increase in the SOC storage in the macro-aggregate fraction (macro-AC) (Table 3).

Compared with RT, the macro-ACs were significantly increased with RS, DS, and RS/DS at the 0–30 cm soil layers, i.e., by 38.97%, 40.13%, and 47.19% at the 0–10 cm depth; by 32.40%, 38.09%, and 24.43% at the 10–20 cm depth, and by 11.45%, 34.44%, and 25.84% at the 20–30 cm depth, respectively. Meanwhile, there was a corresponding decrease in the macro-AC storage with the soil layer depth. Compared with RT, the higher macro-AC storage with residue return treatments were caused by higher SOC storage in >5 mm, 5–2 mm, 2–1 mm, 1–0.5 mm, and 0.5–0.25 mm aggregate fractions at the 0–10 cm depth; by >5 mm, 5–2 mm, 2–1 mm, and 1–0.5 mm aggregate fractions at the 10–20 cm depth, and by >5 mm, 5–2 mm, and 2–1 mm aggregate fractions at the 20–30 cm depth, respectively (Table 3).

RT had a significantly higher SOC storage in the soil micro-aggregate fraction (micro-AC) than that of residue return treatments at the 0–30 cm soil depths, whereas there were no significant differences between RT and RS at 20–30 cm soil depths.

3.4.2. TN Storage in Aggregate

The residue return and tillage methods significantly affected the TN storage and its distribution in different soil aggregate fractions (

Table 4. TN storage in various aggregate fractions under different treatments.

Soil Depth (cm)	Treats	TN (t ha−1)	N Distribution in Aggregate (t ha−1)
			>5 mm	5–2 mm	2–1 mm	1–0.5 mm	0. 5–0.25 mm	<0.25 mm	>0.25 mm
0–10	IC	1.53 b	0.10 b	0.19 a	0.20 c	0.34 b	0.20 b	0.50 a	1.03 b
	RT	1.28 c	0.08 b	0.10 b	0.18 c	0.27 c	0.18 b	0.47 a	0.81 c
	RS	1.61 ab	0.13 a	0.20 a	0.32 a	0.39 ab	0.25 a	0.33 b	1.28 a
	DS	1.64 a	0.09 b	0.19 a	0.27 b	0.45 a	0.21 b	0.44 a	1.21 a
	RS/DS	1.65 a	0.14 a	0.20 a	0.31 ab	0.37 b	0.24 a	0.40 ab	1.25 a
10–20	IC	1.38 c	0.08 c	0.15 c	0.20 c	0.30 bc	0.16 ab	0.49 ab	0.89 b
	RT	1.23 d	0.07 c	0.10 d	0.19 c	0.29 c	0.19 a	0.39 b	0.85 b
	RS	1.48 bc	0.12 a	0.22 a	0.24 b	0.29 c	0.16 ab	0.44 ab	1.03 a
	DS	1.59 ab	0.10 bc	0.20 a	0.31 a	0.35 a	0.14 b	0.49 a	1.09 a
	RS/DS	1.62 a	0.11 ab	0.16 b	0.27 b	0.34 ab	0.19 a	0.55 a	1.07 a
20–30	IC	0.99 b	0.07 a	0.15 b	0.17 c	0.30 a	0.15 c	0.15 c	0.84 b
	RT	0.91 c	0.03 d	0.08 c	0.14 d	0.24 b	0.21 b	0.20 bc	0.71 c
	RS	1.03 b	0.04 c	0.17 a	0.20 b	0.23 b	0.15 c	0.23 b	0.79 b
	DS	1.29 a	0.06 ab	0.08 c	0.22 b	0.27 ab	0.30 a	0.37 a	0.93 a
	RS/DS	1.28 a	0.05 bc	0.09 c	0.28 a	0.29 a	0.22 b	0.35 a	0.93 a

IC, initial condition; RT, five consecutive years of rotary tillage with residues removed; RS, five consecutive years of rotary tillage with residues applied; DS, five consecutive years of deep plowing tillage with residues applied, and RS/DS, DS after RS at a two-year interval. Values followed by different letters within a column in the same year are significantly different at p < 0.05.

). Compared with RT, treatments with residue return significantly increased TN storage, i.e., by 25.92%, 28.84%, and 29.53% at the 0–10 cm depths; by 19.67%, 28.65%, and 31.24% at the 10–20 cm depths, and by 12.89%, 42.23%, and 40.85% at the 20–30 cm depths, respectively. There were no significant differences among the treatments with residue return at the 0–10 cm depth, but the TN storage in RS was significantly lower than that in RS/DS and DS at the 10–20 and 20–30 soil depths.

Soil N distributions in soil macro-aggregate (macro-AN) were significantly increased with RS, DS, and RS/DS in the 0–30 cm layers, i.e., by 58.44%, 49.92%, and 55.05% in the 0–10 cm layers; by 21.77%, 29.24%, and 26.23% in the 10–20 cm layers, and by 12.55%, 31.62%, and 32.12% in the 20–30 cm layers, respectively (Table 4). This was mainly due to the significantly higher N distribution in >5 mm, 5–2 mm, 2–1 mm, 1–0.5 mm, and 0.5–0.25 mm aggregate fractions at the 0–10 cm depth; >5 mm, 5–2 mm, 2–1 mm, and 1–0.5 mm aggregate fractions at the 10–20 cm depth, and >5 mm, 5–2 mm, and 2–1 mm aggregate fractions at the 20–30 cm depth, respectively.

There are some differences in the determination results of soil nitrogen in soil micro-aggregates (micro-AN). At the soil depth of 0–10 cm, the nitrate nitrogen in the RT treatment was significantly higher than that in the straw returning treatment, but at the soil depths of 10–20 cm and 20–30 cm, it was significantly lower than in the straw returning treatment.

3.5. Changes of SOC Storage

Compared with the pre-treatment levels, the SOC storage showed obvious evolution during the study period (Figure 6). The RT significantly reduced the SOC storage by 0.32–1.50 t ha⁻¹ in the 0–30 cm soil layer, and it was mainly due to the fact that the amount of decrease in the macro-AC > the increases in micro-AC (Figure 6). In the 0–10 cm soil layer, the RS treatment increased the SOC storage by 0.36 t ha⁻¹; however, negative effects were observed in the 10–20 and 20–30 cm soil layers, which declined by 0.48 and 0.22 t ha⁻¹, respectively. There were some differences in the evolutions of total SOC storage in the RS, i.e., the increases in macro-AC > the decreases in micro-AC at the 0–10 cm depth; the increases in macro-AC < the decreases in micro-AC at the 10–20 cm depth, and the decreases in macro-AC > the increases in micro-AC at the 20–30 cm depth, respectively (Figure 6). Moreover, DS and RS/DS significantly increased the SOC storage in the 0–30 cm soil layers, i.e., by 0.49 and 0.38 t ha⁻¹ in the 0–10 cm layer; by 1.09 and 0.84 t ha⁻¹ in the 10–20 cm soil layer, and by 0.11 and 0.21 t ha⁻¹ in the 20–30 cm layer, respectively, mainly due to the increases in macro-AC > the decreases in micro-AC (Figure 6).

3.6. Evolution of Soil N Storage

As shown in Figure 7, RT treatment significantly reduced the total nitrogen storage, that is, 0.25 t ha⁻¹ in 0–10 cm soil layer; 0.15 t ha⁻¹ in 10–20 cm soil layer; 0.08 t ha⁻¹ in 20–30 cm soil layer, respectively. The decrease was mainly due to the decrease of both macro-AN and micro-AN in 0–10 cm and 10–20 cm soil layers, and the increase of micro-AN < the decrease of macro-AN in 20–30 cm soil layers (Figure 7). In 0–10 cm soil layer, The residue return treatments did not exhibit any significant variance from one another, but in 10–20 cm and 20–30 cm soil layer, the increase of RS was significantly lower than that of DS and RS/DS treatments RS increased the TN storage by 0.04–0.10 t ha⁻¹ in the 0–30 cm soil layers; the increases were due to the increases in macro-AN > the decreases in micro-AN in the 0–10 and 10–20 cm soil layers, and the increases in micro-AN > the decreases in macro-AN in the 20–30 cm soil layer. Compared to the levels of pretreatment, DS and RS/DS prominently extended the TN storage in the 0–30 cm soil layers (Figure 7), i.e., by 0.11–0.31 t ha⁻¹, and 0.12–0.31 t ha⁻¹, respectively, and the enhancements were due to both the increases in macro-AN and micro-AN in the 0–30 cm soil layer (Figure 7).

3.7. Grain Yield

Combined analysis of variance for that year showed that tillage/residue return treatment and their interactions had significant effects on grain yield (

Table 5. Grain yield of different treatments.

Treats	Grain Yield (t ha−1)
	2016–2017	2017–2018	2018–2019	2019–2020	2020–2021
RT	7.52 b	7.66 b	7.27 c	7.17 c	7.04 c
RS	8.48 a	8.65 a	8.04 b	7.95 b	7.89 b
DS	8.19 a	8.65 a	8.88 a	9.05 a	9.25 a
RS/DS	7.97 ab	8.13 ab	8.98 a	8.99 a	9.20 a
p-value	p (Y)	p (T)	p (Y × T)
	0.0001	0.0001	0.0001

RT, five consecutive years of rotary tillage with residues removed; RS, five consecutive years of rotary tillage with residues applied; DS, five consecutive years of deep plowing tillage with residues applied, and RS/DS, DS after RS at a two-year interval. p value (p) for the effects of year (Y), residue treatments (T) and their interactions (Y × T) are shown. Values followed by different letters within a column in the same year are significantly different at p < 0.05.

).The grain yields for RS, DS, and RS/DS increased by 11.83%, 20.30%, and 18.30% compared with RT, respectively. There were no significant differences among treatments with residue return in the years 2016–2017 and 2017–2018; however, the grain yields for DS and RS/DS were significantly higher than those of RS during the last three years, i.e., by 10.45% and 11.69% in 2018–2019, by 13.80% and 13.00% in 2019–2020, and 17.29% and 16.61% in 2020–2021, respectively. A positive trend in grain yield was noted under both DS and RS/DS conditions during the study period of 5 years, and the average annual growth rates were 3.23% and 3.86%, respectively. A decreasing tendency presented in the continuous rotary treatments (RT and RS) in the last three growing seasons, and the average annual rates were −1.58% and −0.93%, respectively.

4. Discussion

The results of our study demonstrated that the SBD in the soil layer of 0–30 cm treated with residue return significantly decreased, while that in the straw not returning condition increased. This could be due to the increase in SOC, and there was a negative correlation between SBD and SOC content [14,33]. Previous research has demonstrated that the practice of returning long-term crop residues to the field has a beneficial impact on the soil’s overall porosity [34]. Our findings corroborate this, with the residue return treatments notably enhancing the total porosity within the top 0–30 cm soil layer. This improvement was primarily attributed to the integration of corn residues, which facilitates the binding of fine soil particles into larger aggregates. Consequently, the soil’s bulk density was diminished, leading to an increase in total porosity [35].

Soil aggregates serve as the foundation for establishing an optimal soil structure [36]. They are essential for achieving high crop yields [30]. However, traditional tillage practices can disrupt the soil, exacerbating the impact of drying and rewetting cycles as well as freezing and thawing processes, thereby increasing the vulnerability of macro-aggregates to disintegration [37]. Our research indicates that the practice of returning crop residues to the field results in a substantial increase in the quantity and size of soil aggregates compared to the conventional tillage (RT) method. This approach not only diminishes the rate at which macro-aggregates are broken down in agricultural soils, but also enhances the sequestration of organic carbon within micro-aggregates. The micro-aggregates, as a result, become more effective at preserving organic carbon through physical protection, which in turn contributes to the formation of additional macro-aggregates [38]. The MWD and GMD both showed a significant increase following the incorporation of maize residues. Accordingly, it can be inferred that the addition of maize residues exerted beneficial impacts on soil aggregation. This might be attributed to the significant increase in the SOC content after residue incorporation [39]. Furthermore, conventional farming practices can disrupt fungal mycelial networks through the mechanical breakdown of larger soil clumps, subsequently releasing particular organic compounds from these aggregates [40], resulting in the decrease in aggregate stability.

Agricultural measures, including tillage practices and residue returns, not only determine soil physical properties but also affect the root growth, ultimately affecting the land productivity [41]. Our research has demonstrated that the practice of returning crop residues to the field led to a significant enhancement in root length density (RLD) within the top 0–30 cm soil layer. Within the framework of an agricultural ecosystem, the incorporation of these residues facilitated the development of an intricate decomposition subsystem in the upper soil layer, emulating the dynamics of a natural ecosystem. This subsystem serves to mitigate the effects of external forces on the soil matrix and to accumulate both matter and energy in the topsoil, thereby fostering an environment conducive to root growth [37]. Tillage practices resulted in a different spatial distribution of roots, and this might be because of the higher soil moisture content in the top 0–10 cm soil layer as a result of the residue return and rotary tillage [42]. Elevated moisture levels in the upper soil layers fostered root development in those regions, whereas arid soil conditions encouraged the expansion of roots into deeper soil layers [43]. Concurrently, the presence of high mechanical impedance impeded the roots’ ability to penetrate the deeper soil layers [44], a finding that aligns with the results of our study.

Numerous investigations have confirmed that the removal of crop residues can lead to a significant depletion of soil organic carbon (SOC) within agricultural ecosystems [45]. In the present study, generally, significantly higher SOC storage was observed with the residue return treatments, indicating that residue return promoted the SOC sequestrated in the cultivated soil [8,34]. Typically, the accumulation of soil organic carbon (SOC) in agricultural lands is determined by the equilibrium between the addition of crop residues and the decomposition of both fresh and aged soil organic matter (SOM) [37]. Consequently, the practice of returning these residues to the soil can mitigate or counteract the SOC losses that result from intensive and prolonged cultivation practices [46]. There may be two reasons for this: firstly, residue return significantly increased the C input of traditional tillage methods to surface soil [47]; secondly, returning residues to the field promotes soil agglomeration and protects soil organic matter from decomposition. Our results showed that residue return significantly increased the proportion of macro-aggregates and organic carbon storage in macro-aggregates, indicating a synergistic relationship between soil aggregates and soil organic carbon accumulation [30]. Our results also showed that straw return under any tillage practices in a wheat–maize rotation increased SOC and TN, which suggests that substantial straw return is beneficial to improving soil fertility.

Soil nitrogen is a key index for evaluating soil fertility, and it is regulated by agricultural practices such as tillage methods and fertilizer management [48]. In this study, residue return significantly increased the soil nitrogen storage in the 0–30 cm soil layer, indicating that residue return was conducive to nitrogen accumulation and soil fertility. The additional nitrogen is reintegrated into the soil along with the crop residues, and the integration of these residues could improve the physical and chemical properties within the tillage layer, thereby fostering crop growth and encouraging the return of a greater amount of root residues to the soil [32].

Tillage practices are important agricultural strategies for managing and enhancing soil carbon and nitrogen dynamics [17]. The results of this study showed that RT significantly reduced soil C and N pools in three soil layers during the study period. This may be due to the fact that continuous tillage without residues accelerates the decomposition of soil macro-aggregates, which is not conducive to the retention of soil C and N [49]. The study also showed that there was no significant difference in SOC and TN storage in the 0–10 cm soil layer among the RS, DS, and RS/DS treatments, but SOC and TN storage in the 10–20 and 20–30 cm soil layers of the RS treatment were significantly lower than those of the DS treatment, indicating that tillage measures affected the stratification of soil organic carbon (SOC) and total nitrogen (TN) in the soil profile. Changes in soil conditions, such as soil bulk density and total porosity caused by deep tillage and rotary tillage, affect the decomposition rate of straw [50], resulting in differences in soil nutrient accumulation [51]. In addition, deep plowing tillage mixes crop residues and base fertilizer into deeper soil layers [22], which promotes soil agglomeration at different depths and is conducive to soil C and N accumulation.

Previous research has demonstrated that applying crop residues positively influenced both crop yields and the soil’s ability to sustain productivity [52]. These benefits were primarily due to enhancements in the soil’s physical and chemical properties. Mele and Crowley [53] found that traditional tillage practices, which involve the removal or burning of crop residues, can cause significant compaction of the topsoil layers. This compaction creates unfavorable conditions for crop growth, potentially leading to decreased yields. Our findings indicated a substantial increase in wheat yields with the application of residue return treatments as opposed to residue removal. Similar outcomes have been documented in prior studies [27]. An obvious reduction in grain yield was detected over the last three years for the RT and RS treatments. This decrease was linked to the inappropriate soil structure at the 20–30 cm depth, and uneven distribution of nutrients in soil due to the five years of rotary tillage [30]. Contrarily, the DS and RS/DS treatments showed a relatively stable and increasing trend in yield, suggested that combining returning crop straw to the field with deep plowing tillage is beneficial for increasing crop yield. The DS and RS/DS treatments not only improved the soil fertility and soil physical properties but also equilibrated their distribution in the plowing layer, promoted the growth of roots, and ultimately determined land productivity. The year, and the interaction between year and residue management, had a significant effect on grain yield, indicating that the management years of crop residues influenced grain yield, and this effect was attributed to gradual changes in soil quality [37]. From the long-term experimental results, simple rotary tillage operation is not conducive to soil structure, while the combination of rotary tillage and deep tillage operation is more conducive to the improvement of soil structure, enhancing soil fertility and providing a stable foundation for high crop yields.

5. Conclusions

Residue returning can effectively improve the physical and chemical properties of soil. In the five-year field experiment conducted in this study, we observed that residue treatment significantly reduced SBD, increased the total porosity of topsoil (0–30 cm), and greatly improved the soil structure and aggregate stability. The trends in soil organic carbon (SOC) and total nitrogen (TN) storage were similar, but the increase was mainly attributed to the increase in organic carbon (SOC) in the macro-aggregates. During the five-year study period, the DS and RS/DS treatments significantly increased the SOC and TN stocks, especially in the 10–20 and 20–30 cm soil layers, and the increase was significantly higher than that in the RS treatment. In summary, the RS/DS treatment significantly increased the amount of SOC and TN, improved the particle size distribution of the soil aggregates, and improved the soil’s physicochemical properties, which is beneficial for to achieving high wheat yields. Based on a systematic evaluation of key indicators such as SOC, TN, and crop productivity, changing the single tillage mode to a combination of rotary tillage and deep tillage under the condition of returning straw to the field is recommended, as this is conducive to the annual productivity and ecological efficiency of wheat and corn in coordination.