Rational Maize–Soybean Strip Intercropping Planting System Improves Interspecific Relationships and Increases Crop Yield and Income in the China Hexi Oasis Irrigation Area

Abstract

Abundant light and heat in the Hexi Oasis Irrigation Area in China provide superior natural conditions for agricultural development. To study the maize–soybean planting system of intercropping and determine superior group yield and economic benefits in the Hexi Oasis Irrigation Area, eight treatments were set up in 2022–2023: maize–soybean intercropping with a bandwidth of 1.8 m and a row ratio of 2:3 (M1S3), a bandwidth of 1.8 m and a row ratio of 2:4 (M1S4), a bandwidth of 2.0 m and a row ratio of 2:3 (M2S3), a bandwidth of 2.0 m and a row ratio of 2:4 (M2S4), a bandwidth of 2.2 m and a row ratio of 2:3 (M3S3), a bandwidth of 2.2 m and a row ratio of 2:4 (M3S4), monocropping maize (M), and monocropping soybean (S). We analyzed the effects of changes in bandwidth–row ratios on photosynthetic characteristics, yield, and interspecific relationships in these treatments during two crop reproductive periods. Our results showed the following: (1) Under the intercropping system, the photosynthetic capacity of maize was highest when the row ratio was 2∶3 and bandwidth was 1.8 m. The net photosynthetic rate (Pn) increased by 1.72% to 48.90%, the transpiration rate (Tr) increased by 5.53% to 118.10%, and stomatal conductance (Gs) increased by 2.82% to 86.49% compared with other planting systems. Increasing the bandwidth from 1.8 m to 2.2 m improved the photosynthetic characteristics of soybean, increasing Pn, Tr, and Gs by 3.44% to 74.21%, 3.92% to 53.69%, and 2.41% to 55.22%, respectively. (2) The yield of maize and soybean under monocropping was significantly higher than that under intercropping. In the intercropping treatments, the average yield of crops in the M3S3 system was 16,519.4 kg ha⁻¹, an increase of 6.48% compared with the M3S4 system, indicating that the reduction of one row of soybean in the same bandwidth system increases crop yield; The average economic benefit of the M3S3 system over two years was 35,171.73 CNY ha⁻¹, which increased by 13.3 and 80% compared with the average economic benefit of maize and soybean monocropping, indicating that the intercropping system leads to better economic results for farmers than monocropping. (3) In the two-year experiment, the land equivalent ratio (LER) was highest in the M3S3 model, averaging 1.25 over the two years, showing better land productivity compared with other intercropping systems. (4) When bandwidth was 1.8, 2.0, or 2.2 m, the LER decreased by 8.3, 5.9, and 5.6% when planting an additional row of soybeans, the relative crowding coefficient of soybeans in the respective bandwidths increased by 4.59, 4.72, and 0.75%, the competition ratio of maize (CR_M) increased by 22.94, 16.97, and 12.74%, the competition ratio of soybean (CR_S) decreased by 20.47, 17.61, and 16.78%, and the competitive power of maize was greater than that of soybean, indicating that the increase in soybean rows in the same bandwidth system would weaken the competitive advantage of soybean, resulting in crop yield and economic benefit reduction. When the row ratio was 2:3 or 2:4, bandwidth increased from 1.8 m to 2.2 m, LER decreased by 3.31 and 0.86%, intercropping maize aggressiveness (A_M) decreased by 7.55 and 12.50%, CR_M decreased by 18.04 and 24.84%, and CR_S increased by 17.32 and 22.77%, respectively, which indicated that the increase in bandwidth under different row ratio systems could improve the competitive advantage of intercropping soybean, thereby improving crop yield and economic benefits. (5) The AHP method, entropy weight method, and TOPSIS analysis showed that M3S3 ranked first, with the highest comprehensive evaluation (0.6017). In conclusion, the M3S3 planting system can better coordinate crop interspecies relationships, with higher land yield and economic benefit, and can be used as a suitable maize–soybean intercropping system in the Hexi oasis irrigation area.

1. Introduction

The improvement of grain production capacity and output is an important foundation for the sustainable development of China’s economy and society and is the key to guaranteeing national grain security. China’s grain security cannot be separated from that of the world’s, and the world’s grain security cannot be separated from that of China’s [1,2]. The Hexi Corridor is not only an important ecological barrier in northwest China [3] but also an important commodity grain base in China [4], producing about 40% of maize and wheat on 18% of the arable land in Gansu Province [5]. Abundant photothermal resources in the region ensure superior environmental conditions for agricultural production. However, with the comprehensive green transformation of ecological protection and development, the systematic management of mountains, rivers, forests, fields, lakes, grasses, sand, and ice, and the synergistic efficiency of pollution reduction and carbon reduction, production factors such as water, land, and labor force are gradually transferred to non-agricultural industries [6]. The sustainable development of agriculture faces constraints on resources, including water scarcity [7], low mechanization levels [8], low water and fertilizer utilization efficiency [9], and unreasonable agricultural layouts [10]. At the same time, ecological and environmental problems are prominent in agriculture, including pest and weed infestations [11], long-term and large-scale use of pesticides [12], soil drying [13], increasing salinization [14], decreasing biodiversity [15], over-exploitation of groundwater [16], and fragile agro-ecosystems [17]. Additionally, the demand for construction land in arid areas along the Silk Road is increasing, creating competing needs for arable land [6] and restricting sustainable production capacity of oasis agriculture [9]. Therefore, increasing the scientific understanding of the relationship between urban expansion and arable land in the Hexi Corridor as well as improvement of the efficiency of arable land resource utilization are of great significance for realizing a sustainable grain increase [18].

Intercropping has a long history in Chinese grain production. Intercropping planting systems can largely alleviate competition between urbanization and arable land, provide opportunities for sustainable intensification of agriculture, and play an important role in increasing production and efficiency in agro-ecosystems [19]. Intercropping utilizes the complementary advantages of different crops in terms of spatial distribution and nutrient demand [20], promotes the diversification of plant communities in farmland ecosystems [21], improves the replanting index of arable land, alleviates environmental pollution, increases grain production, and improves farmer income, which plays an important role in the development of modern agriculture [22]. However, traditional intercropping poses difficulties with field configuration as well as crop rotation and stubble reversal, conflicts over land, a lack of synergistic fertilization, pest and weed prevention control techniques, and low and unstable yields [23]. Based on the differences in genetic and biological characteristics of crop density tolerance, the strip intercropping layout maximizes the use of time and space in planting intensification. This cultivation method has prominent marginal advantages, is convenient for mechanical operation, can make full use of water, fertilizer, light, heat, and other resources, and doubles the crop yield per unit area. It is in line with the goals of “high yield, mechanization and sustainability” of modern agriculture [24].

Soybean, an important oil crop, protein source, and animal feed raw material, occupies a key position in China’s grain security [25]. To guarantee the security of grain production, priority is given to the planting of staple crops, such as rice, maize, and wheat, while the soybean planting area is restricted and the yields are low [26]. The maize–soybean strip intercropping planting system is an upgrading of traditional intercropping [27] and can ensure no reductions in the yield of maize and increased harvest of soybeans [28]. The staggered arrangement of maize and soybean in height can effectively maximize crop utilization of light energy and arable land [29], promote crop synergy, double the harvest, improve systematic crop yields, and effectively alleviate the competition for land by grain and oil. It is an effective strategy to achieve sustainable ecological agricultural development and revitalization of the soybean industry [30]. At the same time, maize–soybean strip intercropping planting can improve the nitrogen fixation efficiency of soybean root nodules, reduce the amount of nitrogen fertilizer application, improve nitrogen use efficiency, and help to achieve China’s “dual carbon” goal [31].

In the maize–soybean strip intercropping planting system, the tall maize crop can shade the soybeans, limiting soybean yields [32]; however, low levels of shade stress can instead improve the soybean capacity to utilize weak light, increasing its light conversion efficiency [33]. Therefore, combinations of maize plant types and soybean shade-tolerant varieties, wide and narrow row configurations, and field management techniques play an important role in systematic crop yields [34]. By optimizing bandwidth, row ratio, and row spacing of wide and narrow rows, the competition effect between symbiotic crops can be adjusted [35]. The optimal field configuration of 2.0 m bandwidth and 2:2 row ratio of maize and soybean in the Jiangsu riverine area [36] and Sichuan province [37] improved maize–soybean ventilation capacity, increased the interception of light energy by the middle and lower leaves of maize, increased the area of light exposure, improved the utilization rate of light energy, and exhibited a higher systematic yield than that of the net cropping system. The optimal maize and soybean field configuration for yield increase in Guangdong Province is 2.0 m for bandwidth and 2:4 for row ratio [37]. The strip intercropping planting system with a bandwidth of 2.6 to 2.8 m and maize and soybean row ratio of 2:4, or a bandwidth of 3.8 to 4.0 m and a maize and soybean row ratio of 4:4, can be selected for areas with average soil fertility, shorter frost-free periods, and low irrigation and fertilization rates [38].

Most of the existing studies on the optimal field configuration of maize–soybean strip intercropping planting systems are concentrated in Henan, Sichuan, and other areas, with fewer studies in the Hexi Corridor of northwestern China. This study was conducted to determine optimal bandwidths and row spacing between maize and soybeans and soybeans and soybeans for the unique climatic characteristics and arable land fertility status of the Hexi Corridor. Specifically, we examined the effects of different strip intercropping planting systems on the photosynthetic physiological characteristics, yield, economic benefits, and inter-specific competition of maize and soybean crops in cold and cool irrigation areas using AHP, the entropy weight method, and TOPSIS. Our goal was to provide the optimal field allocation strategies for the maize–soybean strip intercropping planting in the Hexi Corridor as well as theoretical and technological support for the popularization and application of this system.

2. Materials and Methods

2.1. Experimental Site

The experiment was conducted in 2022–2023 at Yimin Irrigation Experimental Station, Minle County, the middle part of the Hexi Corridor, Gansu Province (100°43′ E, 38°39′ N, 1970 m a.s.l.) (Figure 1). The climate in the experimental area is temperate continental. The average annual precipitation is 200 mm, annual evaporation is 1900 mm, average annual temperature is 6.0 °C, and the frost-free period is about 105 d, based on meteorological data of the last 20 years. The soil is a light loam, with a topsoil bulk density of 1.46 g cm⁻³ and maximum field water holding capacity of 24%. The basal fertility status and the effective photosynthesizing radiation of the crops during the symbiosis stage are shown in

Table 1. Agro-chemical properties of study site soils.

Year	pH	Organic Matter (g kg−1)	Available Nitrogen (mg kg−1)	Available Potassium (mg kg−1)	Available Phosphorus (mg kg−1)	Effective Radiation (MJ m−2)
2022	7.4	12.8	55.7	190.2	16.1	2517.26
2023	7.8	13.6	62.8	203.5	15.4	2604.38

, and the precipitation and the average temperature over the experiments are shown in Figure 2.

The experimental maize variety was “Longdan 6”, which is a compact type with a developed root system, drought resistance, and lodging resistance. The soybean variety was “Zhonghuang 30”, which is shade-tolerant. Three bandwidths of 1.8, 2.0, and 2.2 m and two row ratios (corn–soybean) of 2:3 and 2:4 were established in this experiment. At the same time, maize (M) and soybean (S) monoculture planting systems were planted for a reference (Figure 3); the specific planting system and plot area are shown in

Table 2. Experimental design showing bandwidth and row ratio field configuration for different treatments.

Treatment	Band Width (cm)	Row Ratio	Maize		Soybean		Maize–Soybean Space (cm)	Density (×104 Plant hm−2)		Plot Area (m2)
			Row Space (cm)	Plant Space (cm)	Row Space (cm)	Plant Space (cm)		Maize	Soybean
M	—	—	50	25.0	—	—	—	8.0	—	80.0
S	—	—	—	—	30	10	—	—	33.3	72.0
M1S3	180	2:3	40	13.8	30	10	40	8.0	11.1	75.2
M2S3	200	2:3	40	12.5	30	10	50	8.0	10.0	83.2
M3S3	220	2:3	40	11.3	30	10	60	8.0	9.1	91.2
M1S4	180	2:4	40	13.8	20	10	40	8.0	16.7	75.2
M2S4	200	2:4	40	12.5	20	10	50	8.0	15.0	83.2
M3S4	220	2:4	40	11.3	20	10	60	8.0	13.6	91.2

. The experiment was arranged as 3 random blocks with eight treatments each (3 bandwidths × two ratios, plus 2 reference), for a total of 24 plots. Experimental maize and soybeans were planted in the first year on 16 April 2022, with soybeans harvested on 1 September and maize harvested on 27 September; in the second year, maize and soybeans were planted on 21 April 2023, with soybeans harvested on 7 September and maize harvested on 30 September. The irrigation level was determined based on the reference crop evapotranspiration (ET₀), which was calculated based on the Penman–Monteith formula; the crop coefficients of maize [39] and soybean [40] in each stage were taken from the literature, and the amount of fertilizer was determined based on the traditional fertilizer application of maize and soybean in the experimental area. To avoid the influence of water and fertilizer on the intercropping planting system, the amount of water and fertilizer applied to each treatment was calculated based on the actual planting density of maize and soybean, as shown in Figure 4.

2.2. Measurements and Methods

2.2.1. Photosynthetic Rate

Five healthy plants conditions were selected for measurements at the jointing, horn mouth, tasseling, grouting, and maturation stages of maize, and at the third trifoliolate, full blooming, podding, and full-grain stages of soybean. The canopies were divided into upper, middle, and lower layers based on plant height, and the net photosynthetic rate (Pn) was measured for the fully unfolded leaves of each canopy from 9:00 to 11:00 in the morning on a sunny day using a portable photosynthetic system analyzer LI-6400, and photosynthetic parameters such as transpiration rate (Tr) and stomatal conductance (Gs) were obtained at the same time. Light intensity in the leaf chamber was set at 1200, 1000, and 800 µmol/(m² s⁻¹) for the upper, middle, and lower layers, respectively, and the temperature was ambient, with round-trip measurements made in the order of the experimental plot layout.

2.2.2. Determination of Yield

The number of maize and soybean plants in each plot was counted before harvesting, and 20 maize and soybean plants were selected for seed testing after harvesting. Ear length, ear diameter, kernel number per plant, ear weight, 100-grain weight, and dry matter accumulation were measured for maize, and pod number per plant, kernel number per plant, grain weight per plant, 100-grain weight, and dry matter accumulation were measured for soybean. In each plot, the remaining maize and soybean plants were harvested separately, air-dried for yield measurement, and finally converted into hectare yield.

2.2.3. Grain Water Productivity

Water productivity (kg m⁻³) = grain yield/water consumption during the whole growth stage.

Irrigation water productivity (kg m⁻³) = grain yield/irrigation amount during the whole growth stage.

2.2.4. Calculation Method of Competition Index

(1)

Land Equivalent Ratio (LER) is a measure of yield advantage.

LER = LER_M + LER_S

(1)

LER_M = Y_IM/Y_SM

(2)

LER_S = Y_IS/Y_SS

(3)

where LER_M and LER_S represent the land equivalent ratio of intercropping maize and soybean, respectively; Y_IM and Y_IS represent the yield of intercropping maize and soybean, respectively; and Y_SM and Y_SS represent the yield of maize M and soybean S monocropping, respectively, in kg hm⁻². When LER > 1, intercropping has an advantage over monocropping, LER < 1 indicates an intercropping disadvantage.

(2)

The relative crowding coefficient (K) is a measure of the competitive advantage of one crop over another crop in intercropping.

K = K_M × K_S

(4)

K_M = Y_IM × Z_IS/[(Y_SM − Y_IM)Z_IM]

(5)

K_S = Y_IS × Z_IM/[(Y_SS − Y_IS)Z_IS]

(6)

Z_IM = (L_M + L_MS)/L

(7)

Z_IS = (L_S + L_MS)/L

(8)

where K_M and K_S represent the relative crowding coefficient of maize and soybean; Z_IS and Z_IM represent the planting proportions of maize and soybean in the intercropping system, respectively; and L_M, L_S, L_MS, and L represent the maize row spacing, soybean row spacing, maize and soybean spacing, and bandwidth, respectively, in cm. The rest of the symbols have the same significance as in the above formulas. When K_M > K_S, maize is more competitive than soybean in the intercropping system; when K_M < K_S, soybean is more competitive than maize in the intercropping system.

(3)

Aggressiveness (A) indicates the extent to which the yield increase of one crop is greater than the yield increase of the other crop in the intercropping system.

A_M = Y_IMZ_IM/Y_SM − Y_ISZ_IS/Y_SS

(9)

A_S = Y_ISZ_IS/Y_SS − Y_IMZ_IM/Y_SM

(10)

where A_M and A_S indicate intercropping maize and intercropping soybean aggressiveness, respectively, and the rest of the symbols have the same significance as in the above formulas. When A_M = A_S, the competitiveness of the two crops is the same; A_M > 0 indicates that maize has an advantage; A_M < 0, indicates that soybean has an advantage.

(4)

Competition Ratio (CR) is an indicator of competition among crops.

CR_M = (LER_MZ_IM/LER_S)/Z_IS

(11)

CR_S = (LER_SZ_IS/LER_M)/Z_IM

(12)

where CR_M and CR_S represent the competition ratios of maize and soybean, respectively, and the rest of the symbols have the same significance as in the above formulas. When CR_M > 1, the competitive ability of maize in the intercropping system is stronger than that of soybean; when CR_M < 1, the competitive ability of soybean in the intercropping system is stronger than that of maize.

2.3. Data Analysis

Microsoft Excel 2019 (Microsoft Corp., Raymond, Washington, DC, USA) software was used to for initial data checking and calculations. SPSS Statistics 24.0 (IBM, Inc., New York, NY, USA) was used to analyze the variability in data for each treatment, and Origin Pro 8.0 (Origin Lab, Corp., Hampton, MA, USA) software was used for plotting. Yaaph v12.5.7528.33196 (Meta Decision Software Technology Co., Ltd., Corp., Taiyuan, China) software was used to draw the comprehensive analytical hierarchical model of the maize–soybean intercropping system and the weight analysis of each index; Matlab (Version R2023b, MathWorks, Corp., Natick, MA, USA) was used to calculate the weights of portfolio assignment based on game theory and the comprehensive score of TOPSIS.

3. Results

3.1. Effects of Bandwidth–Row Ratio Configuration in Maize–Soybean Strip Intercropping Planting System on Crop Photosynthetic Characteristics

3.1.1. Net Photosynthetic Rate

The trends in photosynthetic rate (Pn) were consistent across treatments in 2022 and 2023, and the Pn of maize increased from the jointing stage to a peak at the tasseling stage and then declined from the grouting stage to maturation (Figure 5). At the jointing stage, the M treatment exhibited the highest Pn, with an increase of 10.3 to 23.0% compared to intercropping planting. The strip intercropping planting increased maize Pn by 1.96% to 30.30%, 5.24% to 33.28%, 4.41% to 46.66%, and 15.85% to 95.05%, respectively, at the horn mouth, tasseling, grouting, and maturation stages over that in M, which indicated that strip intercropping could improve the photosynthetic environment of maize leaves, and the effect was more significant in the maturation stage.

Different bandwidth configurations also had significant effects on maize Pn for the same maize–soybean row ratio, and maize Pn at 1.8 m bandwidth and the jointing to maturation stages was increased by 5.05% to 41.38% and 5.42% to 29.52% compared with 2.0 m and 2.2 m bandwidths; this indicated that reasonable bandwidth configurations could improve maize Pn under the stable row ratio system. Under the same bandwidth configuration, different maize–soybean row ratios also affected maize Pn, and 2:4 exhibited a higher Pn than the 2:3 planting system, with 1.07% to 8.62%, 2.76% to 28.33%, and 5.00% to 21.20% increase in 2:3 compared with 2:4 planting in 1.8, 2.0, and 2.2 m bandwidth configurations. When the bandwidth configuration was the same, maize Pn exhibited a gradually decreasing trend as the row ratio increased. Therefore, bandwidth configuration and row ratio can be optimized for increasing the photosynthesis of maize leaves, increasing Pn.

The trends in soybean Pn were consistent in 2022 and 2023, increasing first and then decreasing with the growth stage and reaching the maximum value at the podding stage (Figure 5). Soybean Pn was lower in intercropping than in monocropping at the third trifoliolate, full blooming, podding, and seed-filling stages, with decreases of 2.2 to 38.6%, 4.3 to 35.2%, 6.4 to 39.5%, and 3.6 to 45.3%, respectively, for the stages, mainly due to the shading effect of maize on soybeans, which limited photosynthesis in soybeans. Different bandwidth configurations within a maize–soybean row ratio also had a significant effect on soybean Pn; the photosynthesis of soybean leaves in the 2.2 m bandwidth was enhanced, and soybean Pn from the third trifoliolate to the seed-filling stages increased by 14.03% to 60.62% and 2.96% to 40.40% compared to that in the 1.8 and 2.0 m bandwidths, indicating that with the increase in bandwidth, soybeans access more light, and Pn increases. Different maize–soybean row ratios for the same bandwidth configuration also had an impact on soybean Pn, in which Pn in each soybean growth stage in the 2:3 planting system was higher than that in the 2:4 planting system. Among them, the 2:3 planting system increased by 4.77% to 32.85% in 1.8 m, 6.30% to 28.30% in 2.0 m, and 4.47% to 9.68% in 2.2 m compared with the planting system of 2:4. Thus, with the same bandwidth, soybean Pn gradually decreased, but with the increase in row ratio, the negative effect of row ratio gradually decreased. In the strip intercropping planting system, M3S3 effectively improved the light-receiving environment of soybeans, increasing canopy light transmittance; this is conducive to soybean leaf photosynthesis and maintaining the stability of Pn.

3.1.2. Crop Transpiration Rate

The trend in transpiration rate (Tr) of maize in each treatment in 2022 and 2023 was consistent, increasing from the jointing stage to a peak at the tasseling stage and declining significantly thereafter (Figure 6). At the jointing stage, Tr was highest in the M treatment, but there was no significant difference between M and M1S3, while M2S3, M3S3, M1S4, M2S4, and M3S4 were significantly lower than M by 11.85% to 23.50%, 15.43% to 33.16%, 6.34% to 12.27%, 12.12% to 28.98%, and 20.94% to 41.78%, respectively. Compared with monocropping, strip intercropping planting increased maize Tr by 8.33% to 42.75%, 2.02% to 52.53%, 15.57% to 108.72%, and 26.19% to 188.46%, respectively, during the horn mouth, tasseling, grouting, and maturation stages, which indicated that strip intercropping planting increased maize Pn as well as Tr, especially during the grouting stage and maturation stage.

Different bandwidth configurations also had significant effects on maize Tr within maize–soybean row ratios. Among them, the 1.8 m bandwidth enhanced transpiration in maize leaves, and maize Tr from the jointing to the maturation stage increased by 6.6 to 8.7% compared to the 2.0 and 2.2 m bandwidths; this indicated that expanding the bandwidth was not conducive to the improvement of maize Tr under the same row ratio system. Different maize–soybean row ratios also affected maize Tr under the same bandwidth configuration. Maize Tr was higher in the 2:3 planting system than in the 2:4 planting system in each growth stage, with 1.8, 2.0, and 2.2 m bandwidth configurations exhibiting an increase in Tr of 0.3 to 7.0%. Thus, with the same bandwidth configuration, Tr in each growth stage of maize showed a decreasing trend with the increase in row ratio. Therefore, the M1S3 planting system is more conducive to the transpiration of maize leaves and increases in Tr.

Soybean Tr exhibited an increasing and then decreasing trend with growth stages and reached the maximum value at the podding stage (Figure 6). Soybean Tr was lower in the strip intercropping planting at the third trifoliolate, full blooming, podding, and seed-filling stages than in the monocropping, with decreases of 2.84% to 43.73%, 5.21% to 41.97%, 7.11% to 42.48%, and 4.55% to 35.03%, which was related to the fact that maize weakened light transmittance by shading soybeans, effectively decreasing soybean Tr. Under the same maize–soybean row ratio, different bandwidth configurations also had significant effects on soybean Tr. Among them, the 2.2 m bandwidth was more conducive to the transpiration of soybean leaves, and soybean Tr from the third trifoliolate to the seed-filling stage increased by 16.6 and 3.8% compared with the 1.8 and 2.0 m bandwidths.

Different maize–soybean row ratios also affected soybean Tr under the same bandwidth configuration. Soybean Tr in each growth stage in the 2:3 planting system was higher than that in the 2:4 planting system. Compared with the 2:4 planting system in bandwidth configurations of 1.8, 2.0, and 2.2 m, soybean Tr increased by 6.35% to 25.55%, 11.01% to 28.18%, and 3.92% to 14.38%. Thus, soybean Tr gradually decreased as the row ratio increased within the bandwidth, but the negative effect of row ratio gradually weakened with the increase in bandwidth. The combination of bandwidth and row ratio of M3S3 was optimal in the strip intercropping planting system, alleviating soybean transpiration due to maize shading.

3.1.3. Crop Stomatal Conductance

Maize stomatal conductance (Gs) varied in a unimodal mode with the progression of growth stages, peaking at the tasseling stage (Figure 7). At the jointing stage, maize Gs was lower in all strip intercropping planting treatments than in the monocropping system and was significantly lower in all but M1S3 than that of M by 9.38% to 22.26%, 28.13% to 35.09%, 7.14% to 19.25%, 16.96% to 26.79%, and 36.16% to 41.51%. Maize Gs increased significantly by 1.81% to 124.18% in strip intercropping planting from the horn mouth to the maturation stage compared to monocropping, indicating that strip intercropping planting was conducive to the improvement of maize Gs.

Different bandwidth configurations within the same maize–soybean row ratio also had a significant effect on maize Gs; the 1.8 m bandwidth exhibited an increase of 6.90% to 91.26% and 7.59% to 62.81% in maize Tr from the jointing to the maturation stage compared with the 2.0 and 2.2 m bandwidths, indicating that under the same row ratio system, increasing the bandwidth reduces the passage of gas exchange between the blade and the outside air, increasing stomatal resistance. Under the same bandwidth configuration, different maize–soybean row ratios also affected maize Gs, and maize Gs was higher in the 2:3 than in the 2:4 planting system in each growth stage, with increases ranging from 2.6 to 6.2%, indicating that the same bandwidth configuration suppressed maize leaf stomatal openings as the row ratio increased. In the strip intercropping planting system, the M1S3 planting system was more effective at promoting maize leaf stomatal opening and ensuring Gs balance.

Soybean Gs exhibited an increasing and then decreasing trend over consecutive growth stages, reaching a maximum at the podding stage (Figure 7). Strip intercropping planting was not conducive to the stomatal opening of soybean leaves compared with the S treatment, and Gs at the third trifoliolate, full blooming, podding, and seed-filling stage was reduced by 6.15% to 42.25%, 4.25% to 33.99%, 4.74% to 26.57%, and 4.03% to 40.32% compared with S, respectively. Within the same maize–soybean row ratio, soybean Gs increased with an increasing bandwidth, and the increase with the 2.2 m bandwidth was largest at 13.11% to 47.22% and 1.10% to 31.10% over the 1.8 and 2.0 m bandwidths. For the same bandwidth configuration, different maize–soybean row ratios also affected soybean Gs, and Gs was higher in the 2:3 than in the 2:4 planting system at each growth stage, and in the 1.8 and 2.0 m bandwidth configurations, the G_S at 2:3 increased significantly by 8.2 to 13.6% compared to the 2:4 planting system. There was no significant increase in Gs in the 2.2 m bandwidth. These results indicated that soybean Gs was negatively correlated with row ratio when bandwidth remained the same, but the effect of row ratio on soybean Gs gradually weakened as the bandwidth increased. The combination of M3S3 bandwidth and row ratio was best among the strip intercropping planting systems, resulting in higher Gs for soybeans.

3.2. Effects of Bandwidth–Row Ratio Configuration in Maize–Soybean Strip Intercropping Planting System on System Yield and Its Component Factors

Maize yield and yield components were lower in the strip intercropping planting system than in the monocropping system (

Table 3. Effect of different bandwidth configurations on maize component factors and yield.

Year	Treatment	Ear Length (cm)	Ear Diameter (mm)	Kernel Number Per Ear	Ear Weight (g)	100-Grain Weight (g)	Yield (kg hm−2)	Dry Matter Accumulation Per Plant (g)
2022	M	19.8 ± 0.64 a	51.56 ± 1.93 a	700 ± 29.32 a	236.16 ± 10.71 a	39.46 ± 1.44 a	16,887.64 ± 684.94 a	499.29 ± 20.46 a
	M1S3	19.0 ± 0.93 ab	50.58 ± 1.47 a	636 ± 18.22 ab	214.71 ± 8.88 ab	38.30 ± 1.13 a	14,558.03 ± 428.44 b	468.43 ± 22.32 ab
	M2S3	18.0 ± 0.47 ab	49.14 ± 1.71 a	590 ± 22.90 b	187.37 ± 7.49 cd	33.10 ± 1.14 bc	14,308.25 ± 656.82 b	431.58 ± 14.45 b
	M3S3	17.6 ± 0.82 ab	49.16 ± 1.73 a	584 ± 24.94 bc	200.14 ± 8.89 bc	35.65 ± 1.31 ab	14,115.00 ± 491.44 b	442.07 ± 16.99 ab
	M1S4	18.6 ± 0.78 ab	49.48 ± 1.14 a	608 ± 22.72 b	207.78 ± 4.15 bc	35.89 ± 1.06 ab	14,285.84 ± 506.24 b	455.51 ± 11.18 ab
	M2S4	17.1 ± 0.67 b	48.17 ± 2.21 a	564 ± 17.70 bc	182.89 ± 5.96 cd	31.02 ± 1.23 c	13,880.04 ± 545.36 b	418.96 ± 13.50 b
	M3S4	16.5 ± 0.91 b	46.81 ± 1.82 a	516 ± 18.36 c	174.30 ± 7.03 d	32.17 ± 1.27 bc	13,829.91 ± 701.90 b	411.59 ± 20.87 b
2023	M	19.5 ± 0.8 5a	51.16 ± 0.97 a	674 ± 10.28 a	230.60 ± 14.34 a	37.84 ± 1.30 a	16,200.80 ± 614.25 a	480.25 ± 12.42 a
	M1S3	19.1 ± 0.83 a	49.51 ± 1.91 ab	640 ± 19.31 ab	210.89 ± 13.25 ab	36.22 ± 0.92 ab	14,711.35 ± 849.57 ab	448.71 ± 22.57 ab
	M2S3	17.5 ± 0.55 ab	46.74 ± 1.37 ab	586 ± 19.75 bc	178.26 ± 15.05 b	32.74 ± 1.23 c	14,136.87 ± 224.72 b	427.48 ± 12.62 b
	M3S3	16.8 ± 0.88 ab	45.03 ± 1.31 b	556 ± 18.79 c	198.12 ± 13.12 ab	33.56 ± 0.95 bc	14,679.54 ± 491.76 ab	406.58 ± 11.61 b
	M1S4	18.2 ± 0.84 ab	47.05 ± 1.12 ab	592 ± 14.96 bc	205.15 ± 10.93 ab	34.44 ± 0.74 bc	14,163.08 ± 855.79 b	439.62 ± 13.18 ab
	M2S4	17.4 ± 0.67 ab	45.59 ± 1.47 b	574 ± 22.63 c	181.63 ± 7.32 b	29.05 ± 0.82 d	13,542.12 ± 441.86 b	416.94 ± 12.92 b
	M3S4	15.7 ± 0.98 b	45.32 ± 1.38 b	496 ± 16.21 d	170.27 ± 13.33 b	31.25 ± 1.27 cd	13,219.21 ± 448.76 b	397.58 ± 22.15 b
ANOVA
Year (Y)		ns	*	ns	ns	*	ns	ns
Treatment (T)		**	*	***	***	***	**	*
Y × T		ns	ns	ns	ns	ns	ns	ns

Note: Different lowercase letters in the same column indicate significant differences among different treatments (p < 0.05). The *, ** and *** indicate significant differences among different treatments at the levels of p < 0.05, p < 0.01 and p < 0.001, respectively. The ns means not significant at the level of p ≥ 0.05.

), and there was no significant difference between M1S3 and M treatments. In all but the M1S3 strip intercropping planting systems, ear length, ear diameter, kernel number per ear, ear weight, 100-grain weight, and yield and dry matter accumulation were significantly reduced by 6.06 to 19.49%, 4.03 to 11.98%, 12.17 to 26.41%, 11.04 to 26.19%, 8.99 to 23.23%, 9.39 to 18.40%, and 8.46% to 17.56%, respectively. With the increase in bandwidth configuration, ear length and kernel number per ear continued to decrease, while the ear diameter, ear weight, 100-grain weight, yield, and dry matter accumulation decreased first and then increased, but the yield and component factors in 2.2 m were still lower than in the 1.8 m bandwidth.

As the number of soybean rows increased, maize yield and component factors exhibited a decreasing trend; only at the 2.2 m bandwidth, kernel number per ear, ear weight, and 100-grain weight of maize in row ratio 2:3 significantly increased by 12.10% to 13.18%, 14.83% to 16.36%, and 7.39% to 10.82% compared with that in 2:4 planting system. The increase in the number of soybean rows in 1.8 and 2.0 m bandwidths had no significant effect on the number of component factors of maize. In the strip intercropping planting system, M1S3 was more conducive to maize above-ground biomass accumulation and improved ear structure, and exhibited the highest yield and dry matter accumulation, at 14,558.03 to 14711.35 kg hm⁻² and 448.71 to 468.43 g plant⁻¹, respectively.

Effective strip intercropping planting systems improve soybean yield component factors (

Table 4. Effect of different bandwidth configurations on soybean component factors and yield.

Year	Treatment	Pod Number Per Plant	Kernel Number Per Plant	Grain Weight Per Plant (g)	100-Grain Weight (g)	Yield (kg hm−2)	Dry Matter Accumulation Per Plant (g)
2022	S	34.23 ± 1.38 b	95.86 ± 3.71 b	17.69 ± 0.64 c	19.75 ± 0.40 a	5898.08 ± 325.09 a	53.91 ± 2.20 a
	M1S3	32.84 ± 1.79 bc	93.14 ± 3.93 bc	16.18 ± 0.42 cd	18.72 ± 0.78 ab	1797.95 ± 111.74 bc	27.46 ± 0.97 e
	M2S3	40.37 ± 1.94 a	109.75 ± 5.49 a	20.69 ± 0.35 b	19.03 ± 0.75 ab	2069.33 ± 94.35 bc	34.57 ± 1.31 bc
	M3S3	42.71 ± 1.98 a	118.54 ± 5.30 a	24.01 ± 0.93 a	19.36 ± 0.64 ab	2182.63 ± 103.37 b	36.82 ± 1.11 b
	M1S4	24.36 ± 1.24 d	62.28 ± 3.29 e	9.80 ± 0.28 f	16.54 ± 0.71 c	1634.09 ± 80.05 c	25.94 ± 1.03 e
	M2S4	28.56 ± 1.44 cd	76.33 ± 4.9 d	12.98 ± 0.49 e	17.22 ± 0.66 bc	1946.45 ± 105.91 bc	30.08 ± 1.39 de
	M3S4	29.87 ± 1.61 bc	80.59 ± 3.98 cd	14.53 ± 0.55 de	18.15 ± 0.64 abc	1981.10 ± 102.65 bc	31.96 ± 0.91 cd
2023	S	33.86 ± 1.36 b	93.87 ± 3.90 b	16.31 ± 0.62 c	19.38 ± 0.46 a	5437.13 ± 263.35 a	51.73 ± 2.19 a
	M1S3	33.29 ± 1.06 b	91.64 ± 3.14 b	16.59 ± 0.49 c	18.25 ± 0.35 ab	1842.83 ± 121.58 b	27.09 ± 1.08 d
	M2S3	38.95 ± 1.11 a	107.38 ± 2.46 a	19.92 ± 0.57 b	18.71 ± 0.65 ab	1991.62 ± 116.29 b	33.02 ± 1.65 bc
	M3S3	42.08 ± 1.18 a	115.62 ± 2.96 a	22.68 ± 0.89 a	18.96 ± 0.67 ab	2061.53 ± 150.24 b	34.86 ± 1.00 b
	M1S4	25.13 ± 0.93 d	64.85 ± 2.05 d	10.36 ± 0.41 e	16.27 ± 0.72 c	1726.55 ± 105.37 b	26.48 ± 1.06 d
	M2S4	27.94 ± 0.81 cd	74.44 ± 3.57 c	12.60 ± 0.80 d	16.92 ± 0.67 bc	1889.53 ± 143.63 b	28.64 ± 1.05 d
	M3S4	29.17 ± 1.01 c	77.76 ± 3.16 c	14.64 ± 0.56 c	18.59 ± 0.72 ab	1996.93 ± 118.14 b	30.57 ± 0.91 cd
ANOVA
Year (Y)		ns	ns	ns	ns	ns	ns
Treatment (T)		***	***	***	***	***	***
Y × T		ns	ns	ns	ns	ns	ns

Note: Different lowercase letters in the same column indicate significant differences among different treatments (p < 0.05). The *** indicate significant differences among different treatments at the level of p < 0.001. The ns means not significant at the level of p ≥ 0.05.

); in this study, M2S3 and M3S3 treatments significantly increased soybean pod number per plant, kernel number per plant, and grain weight per plant, by 15.03% to 17.94%, 14.31% to 14.44%, 16.96% to 22.13%, and 24.28% to 24.77%, 23.15% to 23.62%, 35.73% to 39.06% compared with that in monocropping S, but there was no significant difference between M1S3 treatment and S; M1S4, M2S4, and M3S4 treatments resulted in significantly lower soybean yield components than S, with a decrease of 10.2 to 44.6%. M1S4 and M2S4 treatments significantly reduced soybean 100-grain weight, respectively by 16.05% to 16.25% and 12.69% to 12.81%, compared with S; there was no significant change in the 100-grain weight of soybean in the other strip intercropping planting systems. The highest soybean yield and dry matter accumulation of 2061.5 to 2182.6 kg hm⁻² and 34.9 to 36.8 g plant⁻¹ was observed in M3S3 followed by M2S3. At the same maize–soybean row ratio, soybean yields and component factors tended to increase as bandwidth increased, but when the ratio of soybean planting increased in the same bandwidth, soybean yield and component factors decreased.

3.3. Effects of Bandwidth–Row Ratio Configuration in Maize–Soybean Strip Intercropping Planting System on System Water Productivity

The strip intercropping planting system significantly reduced maize water productivity (WP) and irrigation water productivity (IP) compared with monocropping (Figure 8), with decreases ranging from 6.34% to 19.32% for WP and from 8.45% to 20.77% for IP. The M1S3 and M2S3 treatments were more conducive to efficient water utilization by maize, and WP and IP were highest in 2022 in the M2S3 treatment, reaching 1.81 and 2.64 kg m⁻³, respectively, and in 2023 in the M1S3 treatment, reaching 1.92 and 2.60 kg m⁻³. As the proportion of soybean planting ratio increased or bandwidth increased, both the WP and IP of maize tended to decrease, but there were no significant differences among planting systems.

Soybean WP decreased significantly by 21.90% to 60.40% in the strip intercropping planting system compared with the S monocropping (Figure 8), but the M3S3 planting system exhibited the highest soybean IP at 2.31 to 2.34 kg m⁻³, followed by M2S3, which increased by 35.26% to 39.16%, 16.76% to 22.29% compared with S, respectively. IP in the M1S3 treatment was not significantly different from S, but in the M1S4, M2S4, and M3S4 treatments, IP was significantly lower than in S, with a decrease of 10.2 to 44.5%. When the maize–soybean row ratio was the same, increasing bandwidth configurations improved the WP and IP of soybean, with the significantly highest values in the 2.2 m bandwidth system at increases of 26.15% to 35.00% for WP and 36.69% to 48.10% for IP as compared to bandwidth 1.8 m; there were no significant differences in soybean WP between the 2.0 m and 2.2 m bandwidth systems, while IP was significantly reduced by 11.8 to 15.8%. At the same bandwidth system, soybean WP and IP tended to decrease with increasing soybean planting proportion. Overall, soybeans in the M3S3 system exhibited higher WP and IP.

3.4. Effects of Bandwidth–Row Ratio Configuration in Maize–Soybean Strip Intercropping Planting on Systematic Interspecific Competition Indices

The land equivalent ratio (LER) and relative crowding coefficient (K) are different under different strip intercropping planting systems, indicating that different field configurations can affect LER and K values (

Table 5. Effect of different bandwidth–row ratio configurations on systematic interspecific competition indices.

Year	Treatment	LER			K			A		CR
		LERM	LERS	LER	KM	KS	K	AM	AS	CRM	CRS
2022	M1S3	0.86 ± 0.02 a	0.31 ± 0.01 bc	1.17 ± 0.03 a	5.66 ± 0.74 a	0.50 ± 0.03 b	2.88 ± 0.50 a	0.26 ± 0.003 ab	−0.26 ± 0.003 ab	3.24 ± 0.11 b	0.60 ± 0.03 b
	M2S3	0.85 ± 0.03 a	0.35 ± 0.03 ab	1.20 ± 0.06 a	5.38 ± 1.09 a	0.62 ± 0.07 ab	3.47 ± 1.03 a	0.24 ± 0.006 ab	−0.24 ± 0.006 ab	2.72 ± 0.11 cd	0.70 ± 0.05 a
	M3S3	0.84 ± 0.03 a	0.37 ± 0.02 a	1.21 ± 0.0 5a	4.95 ± 0.95 a	0.66 ± 0.06 a	3.37 ± 0.87 a	0.23 ± 0.006 b	−0.23 ± 0.006 a	2.51 ± 0.06 d	0.74 ± 0.04 a
	M1S4	0.85 ± 0.01 a	0.28 ± 0.01 c	1.13 ± 0.01 a	4.16 ± 0.27 a	0.51 ± 0.03 b	2.13 ± 0.16 a	0.28 ± 0.007 a	−0.28 ± 0.007 b	4.08 ± 0.17 a	0.47 ± 0.02 c
	M2S4	0.82 ± 0.04 a	0.33 ± 0.01 ab	1.15 ± 0.03 a	4.00 ± 0.92 a	0.64 ± 0.03 ab	2.50 ± 0.48 a	0.25 ± 0.020 ab	−0.25 ± 0.020 ab	3.22 ± 0.23 b	0.57 ± 0.02 b
	M3S4	0.82 ± 0.05 a	0.34 ± 0.01 ab	1.16 ± 0.07 a	4.51 ± 1.37 a	0.63 ± 0.03 ab	2.94 ± 0.99 a	0.25 ± 0.022 ab	−0.25 ± 0.022 ab	3.05 ± 0.13 bc	0.59 ± 0.02 b
2023	M1S3	0.91 ± 0.03 a	0.34 ± 0.015 ab	1.25 ± 0.020 ab	9.93 ± 2.42 a	0.59 ± 0.04 a	5.69 ± 1.21 ab	0.27 ± 0.007 a	−0.27 ± 0.007 a	3.08 ± 0.21 ab	0.67 ± 0.03 abc
	M2S3	0.87 ± 0.03 ab	0.37 ± 0.012 ab	1.24 ± 0.027 ab	7.44 ± 2.35 ab	0.65 ± 0.03 a	4.73 ± 1.29 ab	0.25 ± 0.018 a	−0.25 ± 0.018 a	2.70 ± 0.17 b	0.72 ± 0.02 ab
	M3S3	0.91 ± 0.02 a	0.38 ± 0.015 a	1.29 ± 0.003 a	9.39 ± 2.01ab	0.68 ± 0.05 a	6.20 ± 0.85 a	0.26 ± 0.013 a	−0.26 ± 0.013 a	2.67 ± 0.16 b	0.75 ± 0.03 a
	M1S4	0.87 ± 0.02 ab	0.32 ± 0.022 b	1.19 ± 0.015 b	5.47 ± 1.09 ab	0.63 ± 0.06 a	3.34 ± 0.46 ab	0.28 ± 0.015 a	−0.28 ± 0.015 a	3.69 ± 0.3 0a	0.54 ± 0.04 d
	M2S4	0.84 ± 0.02 ab	0.35 ± 0.019 ab	1.18 ± 0.003 b	4.14 ± 0.66 ab	0.69 ± 0.05 a	2.77 ± 0.20 b	0.25 ± 0.017 a	−0.25 ± 0.017 a	3.12 ± 0.25 ab	0.60 ± 0.03 cd
	M3S4	0.82 ± 0.03 b	0.37 ± 0.003 ab	1.18 ± 0.003 b	3.96 ± 1.04 b	0.72 ± 0.01 a	2.87 ± 0.77 b	0.24 ± 0.015 a	−0.24 ± 0.015 a	2.79 ± 0.11 b	0.65 ± 0.01 bc
ANOVA
Year (Y)		ns	*	*	*	*	**	ns	ns	ns	*
Treatment (T)		ns	**	ns	ns	*	ns	*	*	***	***
Y × T		ns	ns	ns	ns	ns	ns	ns	ns	ns	ns

Note: Different lowercase letters in the same column indicate significant differences among different treatments (p < 0.05). The *, ** and *** indicate significant differences among different treatments at the levels of p < 0.05, p < 0.01 and p < 0.001. The ns means not significant at the level of p ≥ 0.05.

). In the two-year experiment, LER was >1 (with a range of 1.13 to 1.29) across the strip intercropping planting systems, indicating that all strip intercropping planting systems had intercropping advantages. Under the same bandwidth, LER tended to decrease with the increase in soybean planting proportion, but when the bandwidth increased from 1.8 to 2.2 m in the same row ratio, LER increased slightly, but there was no significant difference. There was no significant difference in LER in any of the treatments in 2022, while the highest LER for the M3S3 system in 2023 was 1.29, which was significantly different from the LER in the 2:4 row ratio planting. The trends in K and LER varied, with the largest K in M3S3, and K_M > K_S under all strip intercropping planting systems, indicating that maize was more competitive than soybean and that maize was in a competitively dominant ecological niche. In this experiment, the LER_M, LER_S, LER, K_M, and K tended to decrease with an increasing row ratio configuration in all three bandwidth systems; thus, when bandwidth was increased from 1.8 to 2.0 m in either row ratio, the LER_M and K_M showed a tendency to decrease, and the LER_S and K_S showed a tendency to increase. In both years of this experiment, A_M was positive and highest in the M1S4 treatment, indicating that maize exhibited a competitive advantage in all strip intercropping planting systems, but there was no significant change in A_M regardless of changes in bandwidth and row ratio. CR_M > CR_S in all strip intercropping planting systems in this study, indicating that maize was more competitive than soybean. Meanwhile, when bandwidth remained constant but the row ratio increased, A_M and CR_M tended to increase, and A_S and CR_S tended to decrease; when the row ratio was constant but the bandwidth increased, both A_M and CR_M tended to decrease, and A_S and CR_S tended to increase.

3.5. Effects of Bandwidth–Row Ratio Configuration in Maize–Soybean Strip Intercropping Planting on Systematic Economic Benefit

In this experiment, total input mainly included irrigation water, fertilizer, seed, labor, and mechanical and other costs (

Table 6. Effects of different bandwidth configurations on the input cost (CNY hm⁻²).

Year	Treatment	Irrigation Water Cost	Fertilizer Cost	Seed Cost	Labor Cost	Mechanical Cost	Other Cost	Total Input
2022	M	1466.14	5039.51	1350	3900	2100	780	14,635.65
	S	887.38	3323.33	1056	3900	2100	780	12,046.71
	M1S3	1750.32	6105.48	1686	3900	2100	780	16,321.80
	M2S3	1676.22	5852.90	1662	3900	2100	780	15,971.12
	M3S3	1753.44	6121.13	1638	3900	2100	780	16,292.57
	M1S4	1898.00	6659.37	1806	3900	2100	780	17,143.37
	M2S4	1809.34	6359.61	1782	3900	2100	780	16,730.95
	M3S4	1874.34	6419.24	1758	3900	2100	780	16,831.58
2023	M	1484.60	5039.51	1350	3650	1970	860	14,354.11
	S	849.58	3323.33	1056	3650	1970	860	11,708.90
	M1S3	1756.09	6105.48	1686	3650	1970	860	16,027.58
	M2S3	1682.54	5852.90	1662	3650	1970	860	15,677.44
	M3S3	1762.07	6121.13	1638	3650	1970	860	16,001.20
	M1S4	1897.69	6659.37	1806	3650	1970	860	16,843.06
	M2S4	1809.96	6359.61	1782	3650	1970	860	16,431.57
	M3S4	1877.90	6419.24	1758	3650	1970	860	16,535.14

). The total inputs into the strip intercropping planting system increased from 1323.3 to 2507.7 CNY hm⁻² and 3924.41 to 5134.16 CNY hm⁻² over those in M and S, respectively (

Table 7. Effects of different bandwidth configurations on the economic benefit.

Year	Treatment	Economic Benefit (CNY hm−2)		Net Proceeds (CNY hm−2)	Output–Input Ratio
		Maize	Soybean
2022	M	48,974.16 ± 1986.33 a	—	34,338.51 ± 1986.33 a	3.35 ± 0.14 a
	S	—	31,849.63 ± 1755.46 a	19,802.92 ± 1755.47 b	2.64 ± 0.15 b
	M1S3	42,218.29 ± 1242.46 b	9708.93 ± 603.42 bc	35,605.41 ± 1828.09 a	3.18 ± 0.11 a
	M2S3	41,493.93 ± 1904.77 b	11,174.38 ± 509.51 bc	36,697.19 ± 2243.87 a	3.30 ± 0.14 a
	M3S3	40,933.50 ± 1425.19 b	11,786.20 ± 558.21 b	36,427.13 ± 1958.78 a	3.24 ± 0.12 a
	M1S4	41,428.94 ± 1468.09 b	8824.09 ± 432.26 c	33,109.65 ± 1709.47 a	2.93 ± 0.10 ab
	M2S4	40,252.12 ± 1581.53 b	10,510.83 ± 571.91 bc	34,032.00 ± 1600.23 a	3.03 ± 0.10 ab
	M3S4	40,106.74 ± 2035.51 b	10,697.94 ± 554.33 bc	33,973.10 ± 2373.68 a	3.02 ± 0.14 ab
2023	M	42,122.08 ± 1597.04 a	—	27,767.97 ± 1597.03 a	2.93 ± 0.11 ab
	S	—	30,991.64 ± 1501.10 a	19,282.74 ± 1501.10 b	2.65 ± 0.13 b
	M1S3	38,249.51 ± 2208.89 ab	10,504.13 ± 692.98 b	32,726.06 ± 2688.89 a	3.04 ± 0.17 ab
	M2S3	36,755.86 ± 584.27 b	11,352.23 ± 662.86 b	32,430.66 ± 797.77 a	3.07 ± 0.05 a
	M3S3	38,166.80 ± 1278.58 ab	11,750.72 ± 856.36 b	33,916.32 ± 1978.42 a	3.12 ± 0.12 a
	M1S4	36,824.01 ± 2225.06 b	9841.34 ± 600.62 b	29,822.28 ± 2299.65 a	2.77 ± 0.14 ab
	M2S4	35,209.51 ± 1148.84 b	10,770.32 ± 818.70 b	29,548.26 ± 1659.47 a	2.80 ± 0.10 ab
	M3S4	34,369.95 ± 1166.76 b	11,382.50 ± 673.41 b	29,217.31 ± 1686.75 a	2.77 ± 0.10 ab
ANOVA
Year (Y)		***	ns	**	**
Treatment (T)		**	***	***	**
Y × T		ns	ns	ns	ns

Note: Different lowercase letters in the same column indicate significant differences among different treatments (p < 0.05). The ** and *** indicate significant differences among different treatments at the levels of p < 0.01 and p < 0.001. The ns means not significant at the level of p ≥ 0.05. The — indicates no data here.

). However, the systematic average net proceeds for the two-year experiment increased from 736.9 to 4,118.5 CNY hm⁻² and 11,923.1 to 15,628.9 CNY hm⁻², an increase of 1.91% to 14.11% and 60.93% to 79.92% over that of M and S. The systematic net proceeds tended to decrease with an increase in the proportion of soybeans planted under constant bandwidth, i.e., the net proceeds from the 2:3 maize–soybean row ratio were higher than those from the 2:4 planting system. The M3S3 treatment resulted in the highest net proceeds, followed by M2S3, but there were no significant differences among different planting systems. There was no significant difference in the output–input ratio between strip intercropping planting and M, but the M2S3 and M3S3 treatments had significantly higher yield and output–input ratios than S, by 15.85% to 25.00% and 17.74% to 22.73%, respectively, whereas the other planting systems were not significantly different from the S treatment.

3.6. Correlation Analysis of Indicators under the Maize–Soybean Strip Intercropping Planting System

Different maize–soybean strip intercropping planting systems changed the growth environment of maize and soybean. The correlation between growth indicators of maize and soybean is shown in Figure 9. Maize yield (YM) and dry matter accumulation (DM) were significantly positively correlated (p < 0.05, r = 0.843) but negatively correlated with soybean yield (YS) and dry matter accumulation (DS), indicating that DM had a negative effect on YS and DS accumulation. YS and DS were highly significantly positively correlated (p < 0.01, r = 0.962), and both were significantly positively correlated (p < 0.05) with soybean water productivity (WPS) and soybean irrigation water productivity (IPS), with a correlation coefficient r of 0.893, 0.838, and 0.911, 0.874, respectively; this indicated that with increasing YS, DS also increased significantly and led to a significant increase in WPS and IPS under the strip intercropping planting system. There was a highly significant positive correlation (p < 0.01, r = 0.992) between maize water productivity (WPM) and irrigation water productivity (IPM), as well as between WPS and IPS (p < 0.01, r = 0.993), indicating that the strip intercropping planting system increased WP and IP. There was a highly significant positive correlation (p < 0.01) between net proceeds (NP) and land equivalent ratio (LER) and relative crowding coefficient (K), with a correlation coefficient r of 0.979; there was a highly significant positive correlation (p < 0.01) between LER and K, with a correlation coefficient r of 0.971, indicating that a reasonable strip intercropping planting system can improve NP while improving LER and K.

3.7. Comprehensive Evaluation

3.7.1. A Comprehensive Evaluation Model

A hierarchical model for the comprehensive evaluation of the strip intercropping planting system was established using Yaaph software. The target layer of comprehensive growth indices (C) included three guideline layers: yield index (C₁), water productivity index (C₂), and economic benefit index (C₃); the yield indices included four index layers: maize yield (C₁₁), soybean yield (C₁₂), maize dry matter accumulation (C₁₃), and soybean dry matter accumulation (C₁₄); the water productivity indices included four index layers: maize water productivity (C₂₁), maize irrigation water productivity (C₂₂), soybean water productivity (C₂₃), and soybean irrigation water productivity (C₂₄); and the economic benefit indices included three index layers: economic benefit (C₃₁), net proceeds (C₃₂), and output–input ratio (C₃₃).

(1)

The AHP Method

We used the AHP method to determine the weighted hierarchical model and then the proportion of 1 to 10 scale method to establish the judgment matrix; finally, the consistency of the matrix was tested. The judgment matrices of the comprehensive growth index, yield index, water productivity index, and economic benefit index are as follows:

$$\begin{gathered} C=\left[\begin{array}{ccc}1.0000& 2.5000& 1.0000\\ 0.4000& 1.0000& 0.3333\\ 1.0000& 3.0000& 1.0000\end{array}\right] C_1=\left[\begin{array}{cccc}1.0000& 1.5000& 1.2000& 3.0000\\ 0.6667& 1.0000& 3.0000& 1.6000\\ 0.8333& 0.3333& 1.0000& 1.2000\\ 0.3333& 0.6250& 0.8333& 1.0000\end{array}\right] \\ {C}_2=\left[\begin{array}{cccc}1.0000& 2.0000& 0.9000& 2.5000\\ 0.5000& 1.0000& 0.3333& 1.5000\\ 1.1111& 3.0000& 1.0000& 1.5000\\ 0.4000& 0.6667& 0.6667& 1.0000\end{array}\right] C_3=\left[\begin{array}{ccc}1.0000& 0.5000& 2.0000\\ 2.0000& 1.0000& 2.0000\\ 0.5000& 0.5000& 1.0000\end{array}\right] \end{gathered}$$

The consistency test coefficients C_R of the comprehensive growth index, yield index, water utilization index, and economic benefit index were <0.10, indicating that the consistency test results were satisfactory and that the established judgment matrix was reliable and reasonable (

Table 8. Results of calculating weights based on the AHP hierarchical analysis method.

Hierarchy	Index	Local Weight	Final Weight	Consistency Test Parameter
Target Layer	yield	0.4100	0.4100	CR = 0.0036 < 0.1 λmax = 3.0037
	water productivity	0.1543	0.1543
	economic benefit	0.4357	0.4357
Target Layer 1	Maize yield	0.3550	0.1456	CR = 0.0719 < 0.1 λmax = 4.1919
	Soybean yield	0.3176	0.1302
	Maize dry matter accumulation	0.1809	0.0742
	Soybean dry matter accumulation	0.1465	0.0601
Target Layer 2	Maize water productivity	0.3309	0.0511	CR = 0.0408 < 0.1 λmax = 4.1090
	Maize irrigation water productivity	0.1648	0.0254
	Soybean water productivity	0.3512	0.0542
	Soybean irrigation water productivity	0.1532	0.0236
Target Layer 3	economic benefit	0.3108	0.1354	CR = 0.0516 < 0.1 λmax = 3.0536
	net proceeds	0.4934	0.2150
	output–input ratio	0.1958	0.0853

, where λ_max is the maximum eigenvalue). The results showed that the weights of each index, in descending order, were net proceeds, maize yield, economic benefit, soybean yield, output–input ratio, maize dry matter accumulation, soybean dry matter accumulation, soybean water productivity, maize water productivity, maize irrigation water productivity, and soybean irrigation water productivity.

(2)

The Entropy Weight Method

The entropy weight method was used to assign weights to a single index, and the weights of each index were calculated (

Table 9. Weights of a single index determined by the entropy weight method.

Indices	C11	C12	C13	C14	C21	C22	C23	C24	C31	C32	C33
Weights	0.07714	0.13982	0.07706	0.09196	0.07707	0.07730	0.09477	0.09421	0.07878	0.08011	0.11177

). The weights of the index determined by the entropy weight method were, in descending order, soybean yield, output–input ratio, soybean water productivity, soybean irrigation water productivity, soybean dry matter accumulation, net proceed, economic benefit, maize irrigation water productivity, maize yield, maize water productivity, and maize dry matter accumulation (Table 9).

(3)

Game Theory Combinatorial Empowerment

To improve the reliability of the weight assignment values and to avoid the influence of subjective factors on the evaluation, a basic weight set was constructed based on the two assignment values obtained with the AHP and the entropy weight method $w={\displaystyle \sum _{\mathrm{k}=1}^l{\alpha }_k}\times w_k^T({\alpha }_k>0)$, where a_k is derived from the AHP method and w_k is derived from the entropy weight method.

Based on the weight set model of game theory, the game model was derived $M\mathrm{in}‖{\displaystyle \sum _{j=1}^i{a}_j\times u_j^T-u_i^T}‖$ i = 1, 2. The combination coefficients after normalization of the above equation were obtained using Matlab: a₁ = 0.8837, a₂ = 0.1163. This yields a vector of combined weights: $w^{\ast }={\displaystyle \sum _{k=1}^2{a}_k^{\ast }}\times u_k^T$; final results are presented in

Table 10. The weight of a single index based on combination assignment.

Indices	C11	C12	C13	C14	C21	C22	C23	C24	C31	C32	C33
Weights	0.1376	0.1313	0.0745	0.0638	0.0541	0.0314	0.0589	0.0318	0.1288	0.1993	0.0884

. The weights of the indices, in descending order, were net proceeds, maize yield, soybean yield, economic benefit, output–input ratio, maize dry matter accumulation, soybean dry matter accumulation, soybean water productivity, maize water productivity, soybean irrigation water productivity, and maize irrigation water productivity (Table 10).

3.7.2. Comprehensive Evaluation Based on the TOPSIS Method

Based on the combination assignment TOPSIS method for comprehensive evaluation, the decision matrix was normalized, the weighting matrix was established, and the ideal solution and fit Ci of the evaluation index were calculated (

Table 11. A comprehensive evaluation of different planting systems and their ranking.

Treatment	C11	C12	C13	C14	C21	C22	C23	C24	C31	C32	C33	D+	D−	Ci	Sequence
M	0.4314	0.0000	0.4210	0.0000	0.4240	0.4297	0.0000	0.0000	0.3532	0.3484	0.3703	0.3552	0.2440	0.4072	8
S	0.0000	0.7676	0.0000	0.5734	0.0000	0.0000	0.5619	0.3809	0.2437	0.2193	0.3125	0.2532	0.3487	0.5793	3
M1S3	0.3816	0.2465	0.3942	0.2961	0.3829	0.3826	0.3273	0.3675	0.3781	0.3833	0.3668	0.2125	0.2762	0.5652	4
M2S3	0.3709	0.2750	0.3692	0.3668	0.3829	0.3840	0.3873	0.4548	0.3731	0.3878	0.3762	0.1934	0.2894	0.5994	2
M3S3	0.3754	0.2874	0.3647	0.3890	0.3602	0.3620	0.4201	0.5221	0.3871	0.3946	0.3750	0.1862	0.2986	0.6159	1
M1S4	0.3709	0.2276	0.3847	0.2845	0.3746	0.3723	0.2455	0.2263	0.3619	0.3530	0.3361	0.2317	0.2539	0.5228	7
M2S4	0.3575	0.2598	0.3593	0.3187	0.3726	0.3693	0.2946	0.2868	0.3566	0.3567	0.3444	0.2139	0.2600	0.5486	6
M3S4	0.3527	0.2694	0.3478	0.3394	0.3437	0.3399	0.3219	0.3271	0.3548	0.3545	0.3420	0.2076	0.2605	0.5565	5
S+	0.4314	0.7676	0.4210	0.5734	0.4240	0.4297	0.5619	0.5221	0.3871	0.3946	0.3762	—	—	—	—
S−	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.2437	0.2193	0.3125	—	—	—	—

). The results showed that the M3S3 planting system had the largest degree of fit for the comprehensive indices (0.6159), followed by M2S3 and S treatment, while the M treatment had the lowest degree of fit, indicating that the comprehensive performance was the worst.

4. Discussion

4.1. Effect of Bandwidth–Row Ratio Configuration on Photosynthetic Rate in Maize–Soybean Strip Intercropping Planting System

Light is an important abiotic factor affecting crop growth [41], and it is the initial source of material metabolism and energy conversion in crops [42]. In the maize–soybean strip intercropping planting system, the planting system and the cultivation environment jointly affect the distribution and utilization of light energy, and the advantages and disadvantages of ventilation and light transmission are key to determining the growth and development of the crop and yield [43]. Therefore, the spatial layout can lead to advantages or disadvantages in strip intercropping planting [44]. In this study, we found that maize Pn, Tr, and Gs were lower in the strip intercropping planting system than in the monocropping system only at the jointing stage, and they were higher than in the monocropping system at all other growth stages. These results were similar to those of previous studies, indicating that the advantages of strip intercropping planting for the photosynthetic performance of maize were more obvious in the middle and late growth stages [45]. The reason may be that, in the strip intercropping planting system, reduced plant spacing (to maintain a similar density of maize in intercropping and monocropping) affects ventilation among plants, the shading effect of maize on soybean is weak during the jointing stage, and competition between the two crop species for nutrients and light energy causes a decrease in the efficiency of light energy utilization during the jointing stage. However, the growth and development of soybeans provided a good environment for rhizobial growth, resulting in enhanced nitrogen fixation and supply for both soybeans and maize and improving the efficiency of maize light energy utilization, This is similar to the conclusions of Shen [46] and Li et al. [47]. However, Wu et al. [27], showed that with the increase in bandwidth distance, light transmittance in the middle and lower part of the maize canopy increased, and the interception rate of light energy decreased, resulting in light loss and a decrease in maize Pn, Tr, and Gs. These results differed from those of this study, which may be related to the direction of row configuration, experimental area environment, characteristics of the tested crop, and planting systems. The Pn, Tr, and Gs of soybean were lower than in the S treatment during the entire period of growth, which may be due to the fact that soybean plants are low to the ground in the strip intercropping planting system, where the tall canopy of maize creates some shade, reducing the light energy interception of soybeans and lowering the photosynthetic capacity of strip intercropping planting of soybean compared to a soybean in monoculture. This is consistent with the conclusions of Zhao et al. [48].

The field configuration of the two crops in the intercropping planting group is an important factor affecting the interspecific relationship and the marginal effect, and it controls light conditions of the dwarf crops. Therefore, different bandwidth configurations also have significant effects on soybean Pn, Tr, and Gs. He et al. [49] concluded that a small bandwidth ensures sufficient light and ventilation for intercropping planting maize, while maize roots can extend sufficiently into the soybean strip to absorb nutrients. However, large bandwidth soybean planting limits the shading effect of neighboring maize. In this study, under the 1.8 m bandwidth configuration, soybean was affected by shading of the neighboring maize and by self-shading; therefore, soybean could only utilize weak or low-efficient light, which reduced the photosynthetic capacity of leaves. However, when bandwidth was extended to 2.2 m, spacing between the maize and soybean increased, light interception by the low photosynthetic layer increased, the utilization of light energy by the soybean population was elevated, and the rate of light energy interception improved, resulting in a significant increase in soybean Pn, Tr, and Gs. This was similar to the results of a previous [49] study.

Cui et al. [45] showed that S maintains peak photosynthetic efficiency, but interplanting with maize at increasing row ratios results in a decrease in soybean Pn, Tr, and Gs. They also found that the closer to maize plants, the more severe the effect of maize was, and the 2:3 planting system was relatively weakly affected by the shading effect of maize. The opposite was true in this study; increasing the planting proportion of soybeans under a fixed bandwidth system also caused a decrease in soybean Pn, Tr, and Gs, which may be attributed to a decrease in row spacing and an increase in plant density as well as to the double stress of maize shading. This conclusion was consistent with former conclusions and also confirmed the results of Cheng et al. [50]. In summary, shading by tall plants is the main reason for the decrease in photosynthetic utilization efficiency of dwarf plants. By expanding the bandwidth and increasing the spacing between maize and soybean, the adverse effects of high stalk crop shading can be reduced, and optimal population productivity can be obtained.

4.2. Effect of Bandwidth–Row Ratio Configuration on Crop Yield, Water Productivity, and Economic Benefits in Maize–Soybean Strip Intercropping Planting Systems

Strip intercropping planting has an important role in increasing crop yields and efficiency [51]. In this experiment, the yield and dry matter accumulation in M and S treatments was higher than that in strip intercropping planting, in agreement with the results of previous studies [52]. One reason for this may be that crop spacing in monoculture lowers competition between individual plants compared with strip intercropping planting [19]. Additionally, the larger the photosynthetic area, the greater the yield and dry matter accumulation [53]; while soybean under the strip intercropping planting system is affected by shading by maize, which weakens soybean photosynthetic capacity over the monocropping system, competition for nutrients in the maize–soybean system in the late stages of growth in turn affects the growth and development of maize and reduces the yield and dry matter accumulation [54]. Finally, the strip intercropping planting system maximizes plant density within the belt, allowing accumulation of CO₂, which is not conducive to crop photosynthesis and crop yield [55]. Different row ratio configurations of strip intercropping planting systems have different effects on crop yield and dry matter accumulation, because changing maize or soybean row ratios shifts the spatial ecology of crops, and with it, the canopy microclimate, affecting the amount of dry matter accumulation [56]. In this experiment, planting an additional row of soybeans at a constant bandwidth reduced the yields of both maize and soybean, which was consistent with the results of Li et al. [55] and indicates that increasing the row ratio configuration beyond the appropriate bandwidth weakens the crop yield potential [57]. At the same time, due to the increase in soybean density, the plant produces a shade avoidance response, in which plant height increases, stem diameter decreases, and average node length increases, resulting in an increase in the rate of stem collapse; meanwhile, densification increases intraspecific competition and the self-shading of plants, decreases individual plant biomass, and decreases branch yield [33].

Feng et al. [19] showed that the planting of one additional row of soybeans could increase the amount of nitrogen and phosphorus transferred from the soybean belt to the maize belt, promote the absorption of nitrogen and phosphorus by maize, facilitate photosynthesis and improve dry matter accumulation, and promote the increase of crop yields; such results are opposite to those of this study, and the difference may be related to the differences in experimental design of the row ratios, regional climatic factors, and physiological characteristics of test crops. Different bandwidth configurations had different effects on the yield increase of the crop population. Under the experimental conditions, crop yield increase was higher in the 2:3 row ratio and 2.2 m bandwidth than in the 1.8 and 2.0 m bandwidth system, whereas the crop yield of the 2:4 row ratio and 1.8 m bandwidth system was higher than that of the 2.0 and 2.2 m bandwidth system, which indicates that a strip intercropping planting system can be optimized to promote crop growth and development, improve the population structure, and increase population yield and that there is a corresponding relationship between row ratio allocation and bandwidth. Reducing or exceeding the appropriate bandwidth can reduce crop yield potential. The reason may be that increasing the bandwidth of row ratio 2:3 can alleviate the effect of weak light stress of maize on soybean, which is conducive to the transfer of photosynthesis products to kernels, improvement of the yield components of soybean, and thus improvement of the yield [58]; however, the increase in row ratio to 2:4 caused the increase in the planting density of maize, resulting in increased competition for light, temperature, water, and fertilizer resources as well as a decrease in single-plant productivity, which led to a decrease in population yield [59]. During the two-year experiment, the trend of soybean yield change under the strip intercropping planting system was consistent, while the trend of maize yield change was inconsistent; the reason for maize inconsistency across years may be related to the effects of ecological factors (precipitation, temperature, and solar radiation) during the experimental year on the agronomic traits and yield of maize.

Maize and soybean water productivity under the strip intercropping planting system was lower than that of the monocropping system, but with the increase in bandwidth, the water productivity of soybean increased, while water productivity of maize decreased; these results were consistent with those of Cai et al. [60]. Meanwhile, we also found that a reasonable strip intercropping planting system could increase soybean irrigation water productivity, while maize irrigation water productivity was always lower than that in monocropping. The strip intercropping planting system had no significant effect on maize water productivity and irrigation water productivity, but soybean water productivity and irrigation water productivity were highest in the M3S3 system, which may be due to a better allocation of soil moisture resources, a reduction in competition for water and nutrients between maize and soybean, and an increase in the complementary effect of water in improving soybean water productivity [60].

This experimental study showed that the strip intercropping planting system, despite greater total input costs in the early stage than those of monocropping, resulted in higher comprehensive yields and increased net proceeds for farmers; this was consistent with the results of Norberg [61]. In the strip intercropping planting system, net proceeds were positively correlated with the change in comprehensive crop yield [54]; with constant bandwidth and increasing row ratio, net proceeds declined, while a bandwidth of 2.2 m and row ratio of 2:3 was conducive to improving crop population yield and net proceeds, increasing economic advantage.

4.3. Effect of Bandwidth–Row Ratio Configuration on Interspecific Competitive Relationships in Maize–Soybean Strip Intercropping Planting System

In this experiment, the strip intercropping planting system exhibited LER > 1, indicating that maize–soybean strip intercropping planting is conducive to improving land use efficiency. During the two-year experiment, the 2.2 m and 2:3 bandwidth and row ratio system exhibited the best LER at 1.25, likely due to optimal light conditions for the accumulation of photosynthesis products, consistent with Cheng et al. [33]. The results of this study showed that when the bandwidth was consistent, planting one additional row of soybean decreased LER, which may be due to increased intraspecific competition for lights and nutrients [62]. Crop roots may also produce harmful root secretions that reduce microbial diversity and enzyme activities in the soil and hinder systemic crop growth [63]. This conclusion is contrary to the findings of Zaeem et al. [64], which may be due to different experimental designs or different coupling effects among crops. At row ratio 2:3, increasing bandwidth could improve LER, while at row ratio 2:4, the results of the 2022 experiment indicated that increasing bandwidth could improve LER, but the results of the 2023 experiment indicated that increasing bandwidth reduced LER, which may be because the crop was more sensitive to the effects of resources such as light, heat, and water in the microclimate of the farmland at row ratio 2:4, which showed inter-annual differences.

The K of different strip intercropping planting systems was different, indicating that different field configuration systems affect both the strength and weakness of interspecific competitive ability. Numerous studies have confirmed that maize is more competitive for resources than soybean in maize–soybean intercropping, i.e., K_M > K_S [65]. In this experiment, K_M was greater than K_S in all strip intercropping planting systems, which again confirmed that maize showed a competitive advantage compared to soybean (see Pelech et al.) [66]. There were differences in crop competition under different bandwidth–row ratio configurations. We found that during the advancement of the reproductive stages, the shading effect of maize on soybean increased (Am and CRm increased, As and CRs decreased) while increased soybean planting increased self-shading and decreased soybean growth. An increased bandwidth and constant row ratio decreased the competitive advantage of maize and facilitated soybean growth. The two-year average LER was the highest and K was the largest in the system of 2.2 m bandwidth and 2:3 row ratio, indicating a mutually beneficial relationship. Therefore, reasonable spatial configuration can promote the separation of ecological niches of groups of crops and the efficient use of crop resources to achieve the maximum yield effect.

5. Conclusions

The maize–soybean strip intercropping planting system allowed for complementary utilization of light, heat, water, and fertilizer resources and improved the photosynthetic capacity of maize during the horn mouth to maturation stages, while soybean exhibited a low light use efficiency due to its low ecological niche position. Light distribution in the system can be changed by expanding bandwidth such that the light environment of the maize population is improved, the light transmission is enhanced, and the net photosynthetic rate of soybean leaves is maximized, increasing the accumulation of photosynthetic products. The average land equivalent ratio of the maize–soybean strip intercropping planting system was >1, indicating the advantage of intercropping, and the competitive advantage of maize was stronger than that of soybean. During the two-year experiment, the average yield and net proceeds of the crop population in the 2.2 m bandwidth and 2:3 row ratio were highest, up to 16,519.35 kg hm⁻² and 35,171.73 CNY hm⁻², indicating that optimization of the population configuration is the key to achieving a coordinated and high yield of maize and soybean in the maize–soybean strip intercropping planting system.

The land equivalent ratio, relative crowding coefficient, output–input ratio, crop yield, and net proceeds exhibited a decreasing trend, and the maize aggressiveness and maize competition ratio showed an increasing trend, with increasing row ratio and constant bandwidth. The land equivalent ratio, relative crowding coefficient, maize competition ratio, net proceeds, and output–input ratio showed an increasing trend, maize aggressiveness showed a decreasing trend, and maize yield decreased and soybean yield increased, with the combined yield showing an increasing trend with increasing bandwidth from 1.8 m to 2.2 m and a constant row ratio. A comprehensive evaluation of multiple indicators of the maize–soybean strip intercropping planting system using the AHP, entropy weighting, and TOPSIS methods showed that the M3S3 system was optimal, with an improved light environment and increased dry matter accumulation in soybean, and with high photosynthetic capacity in maize leaves, which was conducive to the distribution of dry matter to kernels. This laid the foundation for high yields in the later stage of maize growth as well as alleviation of competition for water and nutrients among the crop species, improvement in water productivity and irrigation water productivity, and an increase in net proceeds. We conclude that the planting system with a bandwidth of 2.2 m and a row ratio of 2:3 can maximize the advantages of the maize–soybean strip intercropping planting system in the oasis irrigation area of the Hexi Corridor in China.

Although the planting agronomic technology in recent years has been validated, the development of composite cultivation mechanization is relatively slow, and the performance of the existing maize–soybean composite operating machinery is limited and cannot match the needs of the main planting areas. The mechanization of maize–soybean cultivation is a complex project that requires the combination of agricultural machinery and agronomic technology. Starting from cultivation system and geographic environment of the planting area, and considering the stubble clearing, land preparation, fertilizer requirements, and sowing convergence of the operation as links to the integrated consideration of the design and development of matching intelligent machinery, it is imperative to develop green, efficient, and cost-efficient mechanized cultivation operations to accelerate the realization of the agricultural modernization.