Study of Mechanical-Chemical Synergistic Weeding on Characterization of Weed–Soil Complex and Weed Control Efﬁcacy

: Mechanical-chemical synergy has been proven efﬁcient in weed control. However, characterizing the state of the weed–soil complex after mechanical weeding and revealing its effects on subsequent herbicide application is still challenging, which restricts the implementation of this technology. This paper ﬁrst presents a method to characterize the state of the weed–soil complex from the perspectives of the fragmentation and composite characteristics. The regrowth of the weed–soil complex and the effects of complemented herbicide-reduced spraying on weed control efﬁcacy and crop yield were then investigated. The results showed that the typical diameters of the weed–soil complexes were 10.67 cm and 2.82 cm after inter-row hoe shovel and intra-row ﬁnger weeding, respectively. There were mainly two and four weed–soil complex states after inter-row and intra-row weeding, respectively. The regrowth rate corresponding to the weed–soil complex state with the largest component proportion after inter-row and intra-row weeding was 76.91% and 18.37%, respectively. The additional chemical herbicide sprayed on the weed–soil complex signiﬁcantly improved the fresh weight control efﬁcacy of 95.12% for the preposed inter-row mechanical weeding and 138.07% for the preposed intra-row mechanical weeding in the maize silking stage. The maize yield of inter-row mechanical–75% chemical application treatment was 9.27% higher than that of chemical treatment. Mechanical weeding creates a suitable weed–soil complex state for subsequent chemical application and improves the synergistic weeding effect.


Introduction
Weed control is one of the most vital and challenging tasks in crop production [1]. Effective weeding practices promote crop growth and increase productivity per unit area [2], while improper weed management may result in a roughly 32% yield loss [3]. Therefore, various weeding methods have been proposed for weed control in field production, where representative methods include electric shock, foam, laser weeding, and so on.
Considering the economy and convenience, mechanical weeding and chemical weeding are still in widespread use [4]. Mechanical weeding turns/cuts weeds out of the ground to slow down their growth rate via turning, raking, plowing, and other measures. Mechanical treatments can decrease weed densities, but this is often temporary [5]. Repetitive weeding is required for effective weed control [6]. Instead, chemical weeding technology has been widely adopted for its rapid and thorough weeding effect. However, the excessive application of herbicides presents a range of challenges, from herbicide resistance problems to food safety crises [7,8].

Fragmentation Characteristics
The weed-soil complex appears in different states after mechanical weeding (see Figure 1). The inter-row hoe shovel-type weeding component moves forward and breaks the soil into large complexes, as shown in Figure 1A. The intra-row finger-type weeding component rotary moves and breaks the soil block into small complexes, as shown in Figure 1B. The complexes under different weeding components have irregular shapes and different sizes. We use the typical diameter to define the fragmentation characteristics.

Composite Characteristics
The composite characteristics of the weed-soil complex are defined by weed damage and root-soil contact in this study. Weed damage includes complete separation, partial separation, and complete non-separation of weed stems and roots. Similarly, the root-soil contact also includes complete separation, partial separation, and complete non-separation of weed roots and soil. The nine composite states of the broken complex after mechanical weeding are shown in Figure 2.

Composite Characteristics
The composite characteristics of the weed-soil complex are defined by weed damage and root-soil contact in this study. Weed damage includes complete separation, partial separation, and complete non-separation of weed stems and roots. Similarly, the root-soil contact also includes complete separation, partial separation, and complete non-separation of weed roots and soil. The nine composite states of the broken complex after mechanical weeding are shown in Figure 2. The weed stem and roots are completely separated as shown in Figure 2A-C. The separation states of weed roots and soil are divided into three groups, i.e., weed roots with no soil attachment at all (see Figure 2A), part of weed roots with soil attached (see Figure  2B), and weed roots completely enclosed in soil (see Figure 2C). The weed roots broke and the weed stem and roots are partially connected as shown in Figure 2D,F. The separation states of weed roots and soil are divided into three groups, i.e., weed roots with no soil attachment at all (see Figure 2D), part of weed roots with soil attached (see Figure 2E), and weed roots completely enclosed in soil (see Figure 2F). The weed stem and roots are completely inseparable, and the weed plant is intact, as shown in Figure 2G-I. The separation states of weed roots and soil are divided into three groups, i.e., weed roots with no soil attachment at all (see Figure 2G), part of weed roots with soil attached (see Figure  2H), and weed roots completely enclosed in soil (see Figure 2I). Classifications of composite states relating to the weed and soil in the weed-soil complex: (A) Weed stem and roots are completely separated-weed roots with no soil attachment (C-C), (B) weed stem and roots are completely separated-part of weed roots with soil attached (C-P), (C) weed stem and roots are completely separated-weed roots completely enclosed in soil (C-N), (D) weed stem and roots are partially separated-weed roots with no soil attachment (P-C), (E) weed stem and roots are partially separated-part of weed roots with soil attached (P-P), (F) weed stem and roots are partially separated-weed roots completely enclosed in soil (P-N), (G) weed stem and roots are completely inseparable-weed roots with no soil attachment (N-C), (H) weed stem and roots are completely inseparable-part of weed roots with soil attached (N-P), (I) weed stem and roots are completely inseparable-weed roots completely enclosed in soil (N-N).

Experimental Site
The weed stem and roots are completely separated as shown in Figure 2A-C. The separation states of weed roots and soil are divided into three groups, i.e., weed roots with no soil attachment at all (see Figure 2A), part of weed roots with soil attached (see Figure 2B), and weed roots completely enclosed in soil (see Figure 2C). The weed roots broke and the weed stem and roots are partially connected as shown in Figure 2D,F. The separation states of weed roots and soil are divided into three groups, i.e., weed roots with no soil attachment at all (see Figure 2D), part of weed roots with soil attached (see Figure 2E), and weed roots completely enclosed in soil (see Figure 2F). The weed stem and roots are completely inseparable, and the weed plant is intact, as shown in Figure 2G-I. The separation states of weed roots and soil are divided into three groups, i.e., weed roots with no soil attachment at all (see Figure 2G), part of weed roots with soil attached (see Figure 2H), and weed roots completely enclosed in soil (see Figure 2I).

Experimental Site
The experiment site is located in Zhangqiu District, Jinan City, Shandong Province of P.R. China during the period of June to October 2021. The experimental field (medium fertile) has been under a continuous annual wheat-maize crop rotation for many years. The content of soil organic matter was 11.79 g/kg, available phosphorus was 22.33 mg/kg, alkali hydrolyzed nitrogen was 65.08 mg/kg, available potassium was 144.10 mg/kg, and pH value was 8.17. The maize cultivar Denghai 605 was sown in 70 cm and 22.50 cm for row and plant spacing, respectively. The weeding experiment was carried out during the 4-5 leaf stage of maize, which is the critical period for post-seedling weeding. At this time, the soil wet density was 1.42 g/cm 3 and the average soil moisture content was 16.13% and 16.42% in 0-5 cm and 5-10 cm soil layers, respectively. The soil compactness was 0.66 MPa and 0.49 MPa at 5 cm and 10 cm, respectively.

Experimental Design
The experimental design was formulated by an orthogonal experiment, taking the single or synergistic weeding methods and herbicide application ratio into full consideration. The detailed experimental treatments are shown in Table 1. The inter-row hoe shovel type weeding component and intra-row finger type were used for mechanical weeding components, and the experimental weeding machines were shown in Figure 3. The inter-row weeding treatment (T1) and intra-row weeding treatment (T2) were mechanical weeding treatments without herbicide spraying. The corresponding experimental machines of T1 and T2 are shown in Figure 3. The working speed of the weeding machines was 3 km/h. The herbicide treatment (T5) was used only for chemical weeding, and no mechanical weeding was conducted. The post-seedling herbicide used in T5 was nicosulfuron·mesotrione·atrazine 24% oil dispersion agent (Saipu Industrial Co., Ltd., Jinan, China), which was applied via a manual electric sprayer (3WBD-20, Lujia sprayer factory, Taizhou, China). T3 and T4 were the synergistic weeding of mechanical weeding first and then herbicide supplementation. The dosage of chemical herbicide supplementation was 75% of the conventional dosage, i.e., 150 mL of nicosulfuron mesotrione atrazine 24% oil dispersion agent was mixed with 20 L water.
Considering that rainfall has a great impact on the regrowth of the weed-soil complex after mechanical weeding, experiments were repeated three times over a period of one year. The experimental plots were randomly arranged with an area of 40 m 2 (20 m long and 2 m wide). No pre-seedling and post-seedling herbicides were used in any experimental plots. No additional weeding measures were conducted after the experimental treatments. Inter-row 0 T2 Intra-row 0 T3 Inter-row 75 T4 Intra-row 75 T5 / 100

Weed Control Efficacy
Weed control efficacy was analyzed by the plant control efficacy and the fresh weight control efficacy, and the rate of the injured seedlings was also calculated to evaluate the crop damage caused by mechanical weeding treatments.
Five sample areas 1 m long and 0.8 m wide were randomly selected to record the weed number and fresh weight in each experimental plot. The definitions of plant control efficacy and fresh weight control efficacy are given in Equations (1) and (2), respectively: where E N and E W represent the plant control efficacy and fresh weight control efficacy, respectively; N 0 represents the number of weeds in the non-weeding area; N 1 represents the number of weeds in the treatment area; W 0 represents the fresh weight of weeds in the non-weeding area; W 1 represents the fresh weight of weeds in the treatment area. After one weeding test, the number of maize plants that were buried by soil blocks or scratched by weeding mechanisms within the 30 m test length was recorded. The rate of the injured seedlings is noted in Equation (3): where R I represents the rate of the injured seedlings and I 1 represents the number of maize plants that were injured during weeding.

Typical Diameter
After each weeding test, three sample areas 60 cm in length and 100 cm in width were randomly selected in the weeding path. Then, 10 complexes were randomly picked to record the maximum and minimum diameter individually, the average of which was the maximum and minimum diameter of the weed-soil complex. Here, the average of the maximum and the minimum diameter was defined as the typical diameter of the weed-soil complex.

Weed Damage Degree and Root-Soil Damage Degree
After each weeding test, 20 weed-soil complexes were selected continuously in the weeding path and classified according to Figure 2, and then we calculated the proportion of each type of complex. We specifically hypothesized that (i) when the weed roots are enclosed in a complex but a small portion of the root tip protrudes from the edge of the weed-soil complex, the root tip will be ignored and the weed roots are still considered to be completely enclosed in the weed-soil complex, and (ii) when the weed stem and roots are separated while a small portion of the hairy root remains in the soil, the hairy root will be ignored and no weed root is considered in the soil.

Rate of Regrowth Weeds in the Weed-Soil Complex
The 20 complexes selected for the weed damage degree and root-soil damage degree statistics were labeled separately, and the growth status of the weeds attached to each complex was recorded after one week. Weeds that did not wither and continued to grow are defined as regrowth weeds in this study. The number of complexes in which there was a regrowth weed was recorded separately, and then the regrowth rate of weeds was the ratio of the number of complexes in which there was a regrowth weed to the number of complexes.

Crop Yield
The maize yield in different experimental plots was measured at the maturity stage, and reference [16] was used to calculate the yield.

Data Processing and Analysis
SPSS Statistics 23 was used to analyze the variance of the data, and the LSD (Least Significant Difference) test was used to test the significance of the difference between the experimental treatments. Data calculation and mapping were performed with Excel 2021.

Weed Control Efficacy of Mechanical Weeding
The plant and fresh weight control efficacy of T1 were 82.50% and 79.41%, while they were 48.33% and 46.13% for T2. The plant and fresh weight control efficacy of T1 were better than those of T2. The main reason might be the inter-row weeding component used in T1. The inter-row hoe shovel-type weeding component caused a larger disturbed soil area with its large operational width [21]. Meanwhile, the working depth of the hoe shovel-type weeding component was deeper than that of the intra-row finger-type weeding component, which further enhances soil disturbance and increases weed clearance.
The intra-row finger-type weeding component needs to be operated accurately on the side of the crop lines during the weeding duration. The rate of injured seedlings of T2 was 24.72% due to the bad straightness of the planting path and the lack of a path guide system for the weeder. If visual and control systems were equipped, the rate of injured seedlings could be reduced below 10%, as shown in the study of Quan et al. [22]. There was a large soil disturbance caused by the inter-row weeding component. The resulting large soil blocks often bury maize seedlings, thus causing a rate of injured seedlings of 3.61% during the weeding duration. Mechanical weeding enhances crop vitality [23], although it causes some crop damage.
From the perspective of weed control efficacy for single mechanical weeding, inter-row mechanical weeding has better plant and fresh weight control efficiency with less crop damage than intra-row mechanical weeding. The soil disturbance and fragmentation caused by weeding machinery presented various complex states. The distribution of the typical complex diameter after weeding is shown in Figure 4. The bottom hoe shovel squeezes the complex forward, and the squeezed complex exhibits cracks during the inter-row hoe shovel weeding durations. The continuous forward movement of the hoe shovel causes the cracks to grow wider and wider until the weed-soil complex breaks [21]. The typical diameter of the weed-soil complex after the weeding operation of a hoe shovel was 10.67 cm. The size of the weed-soil complex was large, so weed plants and their roots were enclosed in it. The effective weeding area treated by the hoe shovel-type component was between rows, and the weeds near the crop were not effectively removed. Thus, the hoe shovel component is often used as an inter-row weeding component and causes large disturbances between crop rows.  Regarding the weeding process of the finger weeding component, the bottom metal claw rotates under the self-driven power from ground resistance [24]. The shallow complex was broken and thrown under the action of the rotating metal claw. At the same time, the silica gel finger causes further disturbances to the broken complex. The combined actions of the metal claw and silica gel finger aggravated the fragmentation of the weed-soil complex. The typical diameter of the weed-soil complex was 2.82 cm after the weeding operation of the finger. The size of the weed-soil complex was small, so the weed plants and their roots were less enclosed in the complex. The finger-weeding component can effectively remove weeds from the periphery of the crop while weeds between rows are neglected. It is usually used as an intra-row weeding component with little surface soil disturbance generated between crops.

Composite States of Complex
The contact degree between the roots and soil affects the moisture and nutrient absorption capacity of the plant [23,25]. Thus, the composite states of the roots and soil directly affect weed control efficacy. The proportion of different weed-soil composite states after mechanical weeding is shown in Figure 5. The two main composite states after interrow mechanical weeding are shown in Figure 5A, while the four main composite states after intra-row mechanical weeding are shown in Figure 5B. Regarding the weeding process of the finger weeding component, the bottom metal claw rotates under the self-driven power from ground resistance [24]. The shallow complex was broken and thrown under the action of the rotating metal claw. At the same time, the silica gel finger causes further disturbances to the broken complex. The combined actions of the metal claw and silica gel finger aggravated the fragmentation of the weed-soil complex. The typical diameter of the weed-soil complex was 2.82 cm after the weeding operation of the finger. The size of the weed-soil complex was small, so the weed plants and their roots were less enclosed in the complex. The finger-weeding component can effectively remove weeds from the periphery of the crop while weeds between rows are neglected. It is usually used as an intra-row weeding component with little surface soil disturbance generated between crops.

Composite States of Complex
The contact degree between the roots and soil affects the moisture and nutrient absorption capacity of the plant [23,25]. Thus, the composite states of the roots and soil directly affect weed control efficacy. The proportion of different weed-soil composite states after mechanical weeding is shown in Figure 5. The two main composite states after inter-row mechanical weeding are shown in Figure 5A, while the four main composite states after intra-row mechanical weeding are shown in Figure 5B.
There were two main composite states of the complex due to the wide working width and deep working depth after the inter-row hoe shovel weeding of T1. Peripheral regions beyond the reach of the inter-row weeder were not disturbed, so the weeds were not removed from the soil. This resulted in 17.50% of the weed-soil complexes presenting a state of "weed stem and roots are completely inseparable-weed roots completely enclosed in soil (N-N) "(as shown in Figure 2I). The N-N state is a failure of weeding as the weed grows as it is. The remaining 82.50% of the weed-soil complexes were in a P-N state, i.e., "weed stem and roots are partially separated-weed roots completely enclosed in soil" (as shown in Figure 2F). In general, the composite states of the weed-soil complex depend on the peripheral disturbance around weeds caused by the inter-row weeding components. The larger the disturbance range is, the larger the proportion of the P-N state complex is. and their roots were less enclosed in the complex. The finger-weeding component can effectively remove weeds from the periphery of the crop while weeds between rows are neglected. It is usually used as an intra-row weeding component with little surface soil disturbance generated between crops.

Composite States of Complex
The contact degree between the roots and soil affects the moisture and nutrient absorption capacity of the plant [23,25]. Thus, the composite states of the roots and soil directly affect weed control efficacy. The proportion of different weed-soil composite states after mechanical weeding is shown in Figure 5. The two main composite states after interrow mechanical weeding are shown in Figure 5A, while the four main composite states after intra-row mechanical weeding are shown in Figure 5B.  Due to the small working width and shallow rotation, there were four main composite states of the weed-soil complex after intra-row finger weeding of T2. There were approximately 51.67% of weeds that grow outside the effective action area of weeding mechanisms and are undisturbed during the weeding duration. These complexes present an N-N state, that is, "weed stem and roots are completely inseparable-weed roots completely enclosed in soil (N-N)" (as shown in Figure 2I), causing weeding to fail. The weed may wrap around the weeding finger and be pulled out of the soil during rotation. Hence, the weed-soil complex in this condition will present three composite states. The weed will be easily pulled out of the soil when there are many stubbles in the soil, and the weed-soil complex takes the N-C state of "weed stem and roots are completely inseparable-weed roots with no soil attachment" (as shown in Figure 2G). The N-C state accounted for approximately 29.17%. The roots usually carry soil blocks as the soil is slightly wet when weeds are pulled out, and the other parts of the roots remained underground. At this time, the weed-soil complex presents the P-N state of "weed stem and roots are partially separated-weed roots completely enclosed in soil" (as shown in Figure 2F). The P-N state accounted for approximately 15.83%. When the weeds are deeply rooted and the soil is wetter, the weed stems are easily torn off and the roots remain in the soil, thus presenting a C-N state of "weed stem and roots are completely separated-weed roots completely enclosed in soil" (as shown in Figure 2C), and this state accounted for approximately 3.33%. Generally, the composite states of the weed-soil complex depend on the growing depth of the weed root and the soil condition caused by the intra-row weeding component. Moreover, the soil moisture content directly affects weed damage and root-soil contact, i.e., in which forms of weeds are separated from the soil. Therefore, proper soil moisture conditions are important for the formation of the target weed-soil complex state after mechanical weeding in maize fields.

Regrowth of Weeds Attached to the Weed-Soil Complex
Mechanical weeding breaks the soil into complexes, and then the various sizes and composite states of the weed-root complex affect weed growth. The weeds are prone to attaching to the larger complex to obtain nutrients and water continuously, and then the weed continues to grow well. Figure 6 shows the four typical composite states of the weed-soil complex after mechanical weeding. The weed regrowth studied in this section is mainly based on these four states.

Regrowth of Weeds Attached to the Weed-Soil Complex
Mechanical weeding breaks the soil into complexes, and then the various sizes and composite states of the weed-root complex affect weed growth. The weeds are prone to attaching to the larger complex to obtain nutrients and water continuously, and then the weed continues to grow well. Figure 6 shows the four typical composite states of the weed-soil complex after mechanical weeding. The weed regrowth studied in this section is mainly based on these four states. The size of the broken complex is large after the inter-row hoe shovel weeding. In this event, the weed roots are better preserved in the broken complex. The weeds are closely bound to the soil and prone to regrow due to the undamaged water and nutrient absorption capacity of the weed roots. The weed regrowth rate in the P-N composite state was 76.91%. It infers that the weed-soil complex attached to the weed is so large that the weed regrowth rate is high after inter-row mechanical weeding. If we investigate the weeding control efficacy for a long time, the control effect of a single inter-row mechanical weeding operation is not good, and it needs to be combined with other weeding methods or multiple mechanical weeding.
The weed entangles and moves with rotating parts during the operation of intra-row finger weeding, resulting in weeds being pulled out of the soil. Hence, the weed stems and roots are easily damaged, leading to the separation of the weed and the original growing place. At the same time, the relatively thorough fragmentation of the weed-soil complex causes less soil to be carried by the weed roots. Most of the weed roots are exposed to air, and the water and nutrient absorption capacity of the roots will be greatly reduced, resulting in difficulty regarding weed growth. Weeds in the N-C composite state (see Figure 6D) find it difficult to grow because the stem and roots are completely removed from the soil. The water and nutrient absorption pathways are blocked, resulting in a low weed regrowth rate of 18.37% in this state. As a complex state of P-N shown in Figure 6C, one part of the weed root left in the original growing place continues to grow within the soil easily, and the other part of the weed root that was pulled out is attached to a large complex with smooth water and nutrient absorption pathways. The weed roots will continue to grow in both cases, resulting in a high rate of weed regrowth (61.11%). Regarding the composite state of C-N (see Figure 6A), with the complete separation of the weed stem and weeds, all the roots remain in the original growing place, which creates good conditions for the weed roots to grow continually. In this composite state, the regrowth rate of weeds was as high as 75.00%. Various composite states of the complex have various regrowth results. It will be better if the weed stems and roots do not attach to complexes or attach to smaller complexes, as the weed roots will be exposed to the air and lose consistent water and nutrient absorption. This inference can be used to design and optimize the weeding component. In this state, it is difficult for the weed to maintain a growth situation, and its regrowth rate will be reduced.

Weed Control Efficacy in Maize Silking Stage
Fresh weight control efficacy in the maize silking stage under different weeding treatments is shown in Figure 7. The residual weeds are primarily concentrated between crops after T1, and there are many weeds between and in rows after T2. That is, many weeds regrow after mechanical weeding. Thus, the regrowth of weeds reduced the fresh weight control efficacy of T1 from 79.41% to 40.84%, while T2 was reduced from 46.13% to 31.07%. However, the weed control efficacy of inter-row mechanical weeding treatment (T1) is still significantly better than that of intra-row mechanical weeding treatment (T2) (p = 0.001). Inter-row or intra-row mechanical weeding causes surface soil to loosen, w creases the porosity of the surface soil and helps water and nutrients penetrate [26 ever, it also creates good conditions for weeds to regrow. Weeding machines turn weeds out to form various composite states of the complex, and additional he sprayed at this time will enlarge the contact area between the weeds and herbicid method of mechanical weeding followed by chemical application enhances weed efficacy. The fresh weight control efficacy increased significantly by 95.12% for th row mechanical weeding with additional herbicide sprayed. Similarly, the fresh control efficacy increased significantly by 138.07% and reached 73.96% for the in mechanical weeding with additional herbicide sprayed. This infers that the additi plication of chemical herbicides based on a certain complex composite state cre mechanical weeding will improve weed control efficacy and inhibit weed regrowt pared to the single chemical treatment (T5), the weed control efficacy of single c treatment (T5) was 1.42% higher than that of inter-row mechanical treatment ( there was no significant difference (p = 0.662). The weed control efficacy of inter-r chanical-chemical synergistic weeding treatment (T3) was significantly better tha intra-row mechanical-chemical synergistic weeding treatment (T4) (p = 0.036), and sult is consistent with that of Fang et al. [15]. Therefore, the preposed mechanical w not only affects the weed control efficacy of subsequent chemical herbicides, but d mechanical weeding treatments also have different weed control efficacy values t mechanical-chemical synergistic weeding treatments.

Crop Yield
The maize yield under different experimental treatments is shown in Figur chanical weeding can loosen the soil, help crop roots grow deeper, and ultimat mote crop yield [27]. The soil disturbance caused by inter-row mechanical we more obvious than that of the intra-row mechanical weeding. The working widt inter-row mechanical weeding component is also wider than that of the intra-r chanical weeding component. Therefore, the maize yield treated by single inter-r chanical weeding is 3.39% higher than that of the intra-row mechanical weeding difference between them is not significant (p = 0.572). Inter-row or intra-row mechanical weeding causes surface soil to loosen, which increases the porosity of the surface soil and helps water and nutrients penetrate [26]. However, it also creates good conditions for weeds to regrow. Weeding machines turn or pull weeds out to form various composite states of the complex, and additional herbicides sprayed at this time will enlarge the contact area between the weeds and herbicides. The method of mechanical weeding followed by chemical application enhances weed control efficacy. The fresh weight control efficacy increased significantly by 95.12% for the interrow mechanical weeding with additional herbicide sprayed. Similarly, the fresh weight control efficacy increased significantly by 138.07% and reached 73.96% for the intra-row mechanical weeding with additional herbicide sprayed. This infers that the additional application of chemical herbicides based on a certain complex composite state created by mechanical weeding will improve weed control efficacy and inhibit weed regrowth. Compared to the single chemical treatment (T5), the weed control efficacy of single chemical treatment (T5) was 1.42% higher than that of inter-row mechanical treatment (T3), but there was no significant difference (p = 0.662). The weed control efficacy of inter-row mechanical-chemical synergistic weeding treatment (T3) was significantly better than that of intra-row mechanical-chemical synergistic weeding treatment (T4) (p = 0.036), and the result is consistent with that of Fang et al. [15]. Therefore, the preposed mechanical weeding not only affects the weed control efficacy of subsequent chemical herbicides, but different mechanical weeding treatments also have different weed control efficacy values to that of mechanical-chemical synergistic weeding treatments.

Crop Yield
The maize yield under different experimental treatments is shown in Figure 8. Mechanical weeding can loosen the soil, help crop roots grow deeper, and ultimately promote crop yield [27]. The soil disturbance caused by inter-row mechanical weeding is more obvious than that of the intra-row mechanical weeding. The working width of the interrow mechanical weeding component is also wider than that of the intra-row mechanical weeding component. Therefore, the maize yield treated by single inter-row mechanical weeding is 3.39% higher than that of the intra-row mechanical weeding, but the difference between them is not significant (p = 0.572).
Mechanical weeding creates various composite states of the weed-soil complex, and spraying chemical herbicides can help to reduce the weed regrowth rate and create a suitable environment for crop growth. Based on the preposed inter-row or intra-row mechanical weeding, the additional application of herbicides increased maize yield by 17.65% and 4.77%, respectively. Although there is no significant difference in maize yield between two single mechanical weeding treatments, the yield of inter-row mechanicalchemical treatment is significantly higher than that of intra-row mechanical-chemical treatment by 16.10% (p = 0.016). Mechanical weeding creates various composite states of the weed-soil complex, and spraying chemical herbicides can help to reduce the weed regrowth rate and create a suitable environment for crop growth. Based on the preposed inter-row or intra-row mechanical weeding, the additional application of herbicides increased maize yield by 17.65% and 4.77%, respectively. Although there is no significant difference in maize yield between two single mechanical weeding treatments, the yield of inter-row mechanical-chemical treatment is significantly higher than that of intra-row mechanical-chemical treatment by 16.10% (p = 0.016).
The chemical weeding method kills weeds through the physiological inhibition of metabolism while the mechanical weeding method kills weeds through the physical destruction of the plant. The inhibition of weed regrowth by the chemical method is better than that of the mechanical method. Therefore, the yield of a single-chemical weeding treatment (T5) is higher than that of a single mechanical weeding treatment (T1 or T2), which is consistent with the findings of Fang et al. [15]. Regarding mechanical weeding followed by chemical herbicide application treatment (T3 or T4), the proposed mechanical weeding can promote crop growth and chemical application can inhibit weed regrowth. Therefore, crops can grow better under the double action of mechanical and chemical weeding. The yield of intra-row mechanical-chemical synergy (T4) was not higher than that of a single chemical treatment (T5). There may be three reasons for this. First, intrarow mechanical weeding only removes weeds near crops, and the removed weeds regrow again in large numbers. Second, although reduced chemical herbicide was sprayed, the reduced amount is not sufficient to completely inhibit the weeds' regrowth and remove the weed between rows. Third, intra-row mechanical weeding causes some permanent damage to maize, which also caused some yield loss. However, the maize yield when sprayed with a 25% chemical reduction followed by inter-row mechanical weeding (T3) was the highest. Although there was no significant difference in maize yield between T3 and T5 (p = 0.106), the yield of T3 was still 9.27% higher than that of T5.
Weeding machines create various composite states for the weed-soil complex by breaking the soil. Although the rates of weed regrowth were different in various composite states, it created favorable conditions for subsequent herbicide application. Herbicides with a 25% reduction after inter-row mechanical weeding can obtain an acceptable yield, which was no less than the yield of herbicide treatment. Mechanical weeding before herbicide application can reduce the amount of herbicide without affecting the yield [28]. The chemical weeding method kills weeds through the physiological inhibition of metabolism while the mechanical weeding method kills weeds through the physical destruction of the plant. The inhibition of weed regrowth by the chemical method is better than that of the mechanical method. Therefore, the yield of a single-chemical weeding treatment (T5) is higher than that of a single mechanical weeding treatment (T1 or T2), which is consistent with the findings of Fang et al. [15]. Regarding mechanical weeding followed by chemical herbicide application treatment (T3 or T4), the proposed mechanical weeding can promote crop growth and chemical application can inhibit weed regrowth. Therefore, crops can grow better under the double action of mechanical and chemical weeding. The yield of intra-row mechanical-chemical synergy (T4) was not higher than that of a single chemical treatment (T5). There may be three reasons for this. First, intra-row mechanical weeding only removes weeds near crops, and the removed weeds regrow again in large numbers. Second, although reduced chemical herbicide was sprayed, the reduced amount is not sufficient to completely inhibit the weeds' regrowth and remove the weed between rows. Third, intra-row mechanical weeding causes some permanent damage to maize, which also caused some yield loss. However, the maize yield when sprayed with a 25% chemical reduction followed by inter-row mechanical weeding (T3) was the highest. Although there was no significant difference in maize yield between T3 and T5 (p = 0.106), the yield of T3 was still 9.27% higher than that of T5.
Weeding machines create various composite states for the weed-soil complex by breaking the soil. Although the rates of weed regrowth were different in various composite states, it created favorable conditions for subsequent herbicide application. Herbicides with a 25% reduction after inter-row mechanical weeding can obtain an acceptable yield, which was no less than the yield of herbicide treatment. Mechanical weeding before herbicide application can reduce the amount of herbicide without affecting the yield [28].

Conclusions
This study presented a method to classify the composite states of the weed-soil complex based on weed damage (C, P, N) and root-soil contact (C, P, N). The soil fragmentation and weed regrowth rate under inter-row and intra-row mechanical weeding were then investigated. The effects of herbicide-reduced spraying based on various composite states of the weed-soil complex formed by mechanical weeding on weed control efficacy and crop yield were also investigated. The results showed that the states of the broken complex after mechanical weeding can be characterized by weed damage and root-soil contact, which were used to distinguish the different degrees of broken complexes caused by different motions of weeding components. The complex after effective inter-row mechanical weeding primarily presents a P-N state, while the weed-soil complex mainly presents N-C, P-N, and C-N states after effective intra-row mechanical weeding. The weeds located inside the weed-soil complex with different composite states had different weed regrowth rates. The rate of weed regrowth in the P-N state caused by inter-row mechanical weeding was 76.91%, while the regrowth rates in the N-C, P-N, and C-N states caused by intra-row mechanical weeding were 18.37%, 61.11%, and 75.00%, respectively. Due to the additional chemical herbicide sprayed on the various composite states of the complexes, a significant improvement in fresh weight control efficacy of 95.12% for the preposed inter-row mechanical weeding and 138.07% for the preposed intra-row mechanical weeding in the maize silking stage occurred. The maize yield of T3 was 9.27% higher than that of T5. It was the preposed mechanical weeding that creates different composite states of weed-soil complexes through the weeding process.