Effects of Maize Varieties on Biomass Yield and Silage Quality of Maize–Soybean Intercropping in the Qinghai–Tibet Plateau

Abstract

Forage deficiency is the bottleneck that restricts the development of plateau animal husbandry. Maize (Zea mays L.)–soybean (Glycine max L.) intercropping can improve the forage biomass yield and silage quality. This experiment was conducted in Ganzi Tibetan Autonomous Prefecture to explore the effects of four maize varieties (M1, Rongyu Silage No. 1; M2, Yayu 04889; M3, Demeiya No. 1; M4, Zhenghong 505) on biomass yield, nutritional composition, and silage quality in maize–soybean intercropping. The results showed that M1S had the highest total dry matter yield (18.03 t ha⁻¹), M3S had the highest crude protein (CP) content (8.46% DM), and soybeans had the highest water-soluble carbohydrate (WSC) content (8.55% DM). After silage, the CP content (13.44% DM) of mixed silage in M3S was higher, and the contents of neutral detergent fiber (39.42% DM) and acid detergent fiber (25.42% DM) were lower than those in maize silage alone. The WSC content (4.45% DM) of mixed silage in M3S was higher and the pH value (4.46) and ammonia–nitrogen to total nitrogen (3.97%) were lower than those of soybean silage alone. The results of membership function analysis showed that M3S was the best in fresh feeding and silage utilization, followed by M1S. Therefore, M3S (Demeiya No 1. intercropped with soybeans) is recommended in high-altitude areas.

1. Introduction

The yak is one of the main livestock animals in the plateau pastoral area, which is mainly distributed in the Qinghai–Tibet Plateau and the surrounding alpine regions above 3000 m in China [1]. China is the country with the largest number of yaks, accounting for about 92% of the global total [2], mainly distributed in Sichuan, Tibet, Qinghai, and Xinjiang [3]. Sichuan Ganzi is located on the eastern edge of the Qinghai–Tibet Plateau. Animal husbandry is the dominant industry in the region, and yaks are the characteristic livestock; their average annual inventory is more than 2 million, accounting for about 15% of the total number of yaks in the country. Due to the influence of low temperatures, long winters, and extensive management [4], the biomass yield of artificial planted pasture in the Tibetan Plateau is low and cannot meet the needs of yaks for feed in winter and spring. It can be seen that the lack of feed in winter and spring is the bottleneck that restricts the healthy and sustainable development of the yak industry. It has been found that maize–soybean intercropping can not only improve the biomass yield [5] but also preserve the nutrients of forage for a long time via mixed silage [6], effectively solving the seasonal imbalance of forage supply.

Maize (Zea mays L.) is the main source of energy feed for livestock, is rich in water-soluble carbohydrates, and can be used as the main silage material [7], but its crude protein content is low. Soybeans (Glycine max L.) have high crude protein content and are rich in nutrients, but their water-soluble carbohydrate content is low and cannot meet the requirements of lactic acid bacteria for substrates, and their buffer capacity is large [8] and cannot reach low pH levels, so they often fail when used along for silage [9]. The mixed maize–soybean silage can not only improve the fermentation quality but also meet the demands of livestock for nutrition. In recent years, Yang Wenyu’s team created a wide and narrow strip mode of maize–soybean intercropping by improving the traditional intercropping mode [10,11]. In this mode, plants can make full use of light and heat resources by matching height with height [12]. A large number of studies have shown that maize–soybean intercropping can improve the soil fertility [13] and land equivalent ratio [14], increase biomass yield, and improve nutritional quality [15,16], and the quality of the mixed silage after maize–soybean intercropping is better than that of sole silage [17].

It can be seen that maize–soybean intercropping has proven to be an effective way to improve forage yield and silage quality. However, different maize varieties show different effects in intercropping [18]. A study has shown that there is interaction between maize varieties and experimental sites, and the yield of the same maize variety planted in different locations can differ significantly [19]. Gupta et al. found that maize varieties have different effects on plant height and nutrient composition in maize–cowpea intercropping [20]. The experimental results of Yang et al., where three maize varieties and three soybean varieties were intercropped separately, showed that there were differences in the dry matter weight of seeds and the growth parameters among the intercropping combinations [18]. At present, few studies have focused on maize varieties suitable for intercropping with soybeans in high-altitude areas. Therefore, this experiment was carried out in Ganzi. Four high-quality maize varieties with high yield and disease resistance characteristics were selected for intercropping with soybeans to explore the effects of maize varieties on forage yield, nutrient composition, and silage quality in maize–soybean intercropping, so as to obtain maize varieties suitable for intercropping with soybeans in high-altitude areas, and to provide technical support for the development of animal husbandry in this area.

2. Materials and Methods

2.1. Experimental Site

The field experiment was conducted in the town of Bamei (E 101°48′, N 30°48′, average altitude 3500 m), Daofu County, Kangding, Ganzi Tibetan Autonomous Prefecture. Bamei has a cold temperate continental monsoon climate, with an average annual temperature of 8.2 °C, annual precipitation of 668 mm, annual average sunshine of 2296.95 h, and a frost-free period of about 113 d throughout the year. The soil had a pH of 6.8, organic matter of 12.5 g·kg⁻¹, available N of 158.7 mg·kg⁻¹, available P of 65.4 mg·kg⁻¹, and available K of 185.9 mg·kg⁻¹ (0–20 cm topsoil layer). The weather data of the experiment are shown in

Table 1. Monthly minimum temperature, maximum temperature, mean temperature, and rainfall from April to September during the cropping period in 2018.

Month	Minimum T (°C)	Maximum T (°C)	Mean T (°C)	Rainfall (mm)
April	1.2	14.6	7.9	24
May	5.6	18.5	12.0	53
June	9.6	20.2	14.9	104
July	11.2	20.6	15.9	174
August	10.4	20.1	15.3	106
September	9.1	18.0	13.5	132

T: temperature.

.

2.2. Field Experiment and Silage Preparation

The experiment was a single-factor randomized block design. Three plots were set for each treatment, and all plots were within a 1.2 m distance of one another to eliminate disturbance from the nutrient flow. The factor was maize variety, including M1 (Rongyu Silage No. 1, Sichuan Agricultural University, Chengdu, China; growth period: 119 days), M2 (Yayu 04889, Sichuan Yayu Science and Technology Development Co., Ltd., Chengdu, China; growth period: 116 days), M3 (Demeiya No. 1, Heilongjiang Kinfeng Seed Industry Co., Ltd., Harbin, China; growth period: 110 days), and M4 (Zhenghong 505, Sichuan Nongda Zhenghong Seed Industry Co., Ltd., Chengdu, China; growth period: 118 days); these four maize varieties were intercropped with soybeans (S: Nanxiadou 25, Nanchong Academy of Agricultural Sciences, Nanchong, China; growth period: 134 days), namely, M1S, M2S, M3S, and M4S, respectively.

Maize and soybeans were sown on 24 April 2018 as shown in Figure 1; the planting density of maize was 67,500 plants/ha, while that of soybeans was 135,000 plants/ha. Each plot was 7 m long and 5.7 m wide and had three maize–soybean strips; 422 kg·ha⁻¹ pure nitrogen, 169 kg·ha⁻¹ P₂O₅, and 169 kg·ha⁻¹ K₂O were applied to the maize. P₂O₅ and K₂O were applied once as the base fertilizer, while pure nitrogen was applied three times as a base fertilizer, seedling fertilizer, and jointing fertilizer at a ratio of 3:3:4, and the soybeans were not fertilized. The maize and soybeans were harvested at the same time on 15 September 2018 (before the frost). At this time, M1 and M4 were in the late stage of milk ripening, M2 and M3 were in the early stage of milk ripening, and S was in the bulging stage.

Before harvest, we randomly selected a maize–soybean strip in each plot. According to the planting density ratio of maize and soybeans, we took 8 maize plants (4 plants in each row) and 16 soybean plants (8 plants in each row) for plant height measurement, harvested the above plants and measured their yield (the stubble height of harvesting was 20 cm for maize and soybeans), and then cut these into 2–3 cm segments. One part was used to determine the chemical composition of fresh samples before silage, and the other part was put into a polyethylene bag, which was vacuumed and sealed with a vacuum pump (no additives). Each bag was filled with 0.3 kg. In addition, the maize and soybeans in each treatment were ensiled separately. A total of 36 bags were used (4 maize varieties × 3 kinds of silage × 3 replicates). The bags were stored indoors near the test site (4–14 °C). After 60 days, the bags were opened for quality determination.

2.3. Chemical Composition

We took 200 g of the fresh and silage materials, which were killed at 105 °C for 30 min and dried at 65 °C to constant weight, transferred to a dryer, cooled to room temperature, weighed, and their dry matter (DM) contents were calculated, after which the dry samples were ground and sieved (40 mesh, pore size 0.425 mm). The Kjeldahl nitrogen method [21] was used to determine the crude protein (CP) contents in the samples. About 0.2 g of each dried sieved sample was weighed with a 1/10,000 balance (the weight of the samples was recorded) and then put into the Kjeldahl nitrogen digestion tube. One piece of catalyst and 12 mL of concentrated sulfuric acid were added in succession. The forage was boiled in a digestion furnace at 420 °C for 90 min and cooled to room temperature in a ventilator. The CP content of the forage was determined using a FOSS KJelTEC-8400 automatic nitrogen meter. The contents of water-soluble carbohydrate (WSC), neutral detergent fiber (NDF), and acid detergent fiber (ADF) were determined by the anthrone colorimetric method [21] and van Soest’s fiber method [22], respectively. According to the above indices, the relative feeding value (RFV) of the forage was calculated [23].

The calculation formula was as follows:

$$\mathrm{RFV}=\mathrm{DMI}\times \mathrm{DDM}÷1.29$$

$$\mathrm{DMI}=120÷\mathrm{NDF}$$

$$\mathrm{DDM}=88.9-0.779\times \mathrm{ADF}$$

where DMI is the random intake of dry matter of roughage, while DDM is the digestible dry matter.

2.4. pH and NH₃-N

After 60 days of silage, 20 g of sample was weighed, 180 mL of deionized water was added, and the sample was sealed and placed in a 4 °C refrigerator for leaching overnight (>24 h). The treated extract was filtered with 4 layers of gauze, and the filtrate was stored in a −30 °C refrigerator. Part of the filtrate was taken, and the pH of the filtrate was measured with a pH meter (pH S-3C). The content of ammonia nitrogen (NH₃-N) was determined by the phenol–hypochlorous acid method [24] as follows: 20 g of mixed silage was weighed, 180 mL of distilled water was added, and the mixture was stirred in a small juice extractor for 1.5 min before filtration. A certain amount of filtrate was taken, trichloroacetic acid (TCA) was added at a 4:1 ratio, and then the mixture was kept in the refrigerator at 4 °C overnight. We then drew 50 μL of the filtrate (distilled water as the blank) into a labeled test tube, added 50 μL of distilled water and 2.5 mL of phenol reagent successively, shook the tube well, and then added 2 mL of sodium hypochlorite reagent, mixed the contents well, and placed the tube in a 95 °C water bath for 5 min. After cooling, the ammoniacal nitrogen content was calculated by colorimetry at a wavelength of 630 nm.

2.5. Statistical Analysis

Excel 2019 was used for data collation; SPSS22.0 software was used for analysis of variance; the Duncan method (p < 0.05) was used for multiple comparisons. The final data were expressed as the average of three replicates. OriginPro 2022 was used to convert the results into graphs, and the membership function value method in fuzzy mathematics was used to comprehensively evaluate all indicators. The membership function method was used to calculate the membership function value of each index, and then the average was calculated as the average membership degree of the treatment for ranking—the larger the average membership degree, the better the intercropping combination.

The calculation formula was as follows:

$$\mathrm{X}({\mathsf{\mu }}_1)=\frac{\mathrm{X}-{\mathrm{X}}_{\min }}{{\mathrm{X}}_{\max }-{\mathrm{X}}_{\min }}$$

$$\mathrm{X}({\mathsf{\mu }}_2)=1-\mathrm{X}({\mathsf{\mu }}_1)$$

where Xµ is the membership degree of an index under a certain treatment, X is the measured value of the corresponding index of the treatment, X_min and X_max are the minima and maxima of the corresponding indices of all treatments, respectively, X(µ₁) is positively correlated with each index, and X(µ₂) is negatively correlated with each index. NDF, ADF, pH, and NH₃-N/TN are negative membership functions, and the others are positive membership functions.

3. Results

3.1. Effects of Maize Varieties on the Yield and Plant Height in Maize–Soybean Intercropping

As shown in

Table 2. Effects of maize varieties on the yield of maize–soybean intercropping.

Treatments	Fresh Matter Yield (t ha−1)			Dry Matter Yield (t ha−1)
	M	S	MS	M	S	MS
M1S	41.67 a	9.86 a	51.53 a	15.36 a	2.67 a	18.03 a
M2S	46.83 a	3.65 d	50.48 a	10.71 b	0.68 c	11.39 b
M3S	48.70 a	3.89 c	52.59 a	11.04 b	1.25 b	12.29 b
M4S	42.98 a	8.98 b	51.96 a	14.23 a	2.69 a	16.91 a
SEM	1.42	1.42	0.38	1.00	0.44	1.43

M1S, M1 (maize, Rongyu Silage No. 1) and S (soybean, Nanxiadou 25) intercropping; M2S, M2 (maize, Yayu 04889) and S intercropping; M3S, M3 (maize, Demeiya No. 1) and S intercropping; M4S, M4 (maize, Zhenghong 505) and S intercropping; M, average yield of maize in the corresponding treatment; S, average yield of soybeans in the corresponding treatment; MS, average total yield of maize and soybeans in the corresponding treatments; SEM, standard error of the mean. Means in the same column that do not share the same letters differ significantly (p < 0.05).

, the yield of maize–soybean intercropping was significantly affected by the maize variety (p < 0.05). In terms of fresh forage yield, M3 had the highest yield (48.70 t ha⁻¹), but there was no significant difference in maize yield between treatments. The soybean yield of M1S (9.86 t ha⁻¹) was the highest—significantly higher than that of the other treatments (p < 0.05)—followed by M4S. The total fresh forage yield of M3S (52.59 t ha⁻¹) was the highest, but there was no significant difference between the treatments. In terms of dry matter yield, M1 (15.36 t ha⁻¹) had the highest yield, followed by M4, while that of M2 was the lowest. The soybean yield of M4S (2.69 t ha⁻¹) was the highest, followed by M1S, while the lowest was M2S, and the total dry matter yield of M1S (18.03 t ha⁻¹) was the highest—significantly higher than that of M2S and M3S (p < 0.05).

As shown in Figure 2, the plant height of M4 was the highest (314.83 cm), which was significantly higher than that in the other treatments (p < 0.05), while that of M3 was the lowest. The plant height of soybeans in M1S was the highest (135.00 cm), followed by M2S (121.67 cm), while that of M3S was the lowest (60.89 cm).

3.2. Effects of Maize Varieties on the Chemical Composition of Maize and Soybeans in Maize–Soybean Intercropping

As shown in

Table 3. Effects of maize varieties on the chemical composition of maize in maize–soybean intercropping.

Treatments	DM%	CP/%DM	WSC/%DM	NDF/%DM	ADF/%DM	RFV
M1S	36.86 a	6.52 c	10.40 c	50.42 c	30.60 b	120.04 a
M2S	22.88 c	7.54 b	13.19 b	55.55 a	35.00 a	103.26 b
M3S	22.68 d	8.46 a	14.59 a	52.21 b	27.45 c	120.32 a
M4S	33.10 b	6.81 c	10.77 c	49.60 c	29.25 bc	124.00 a
SEM	3.12	0.37	0.86	1.14	1.39	4.02

M1S, M1 (maize, Rongyu Silage No. 1) and S (soybean, Nanxiadou 25) intercropping; M2S, M2 (maize, Yayu 04889) and S intercropping; M3S, M3 (maize, Demeiya No. 1) and S intercropping; M4S, M4 (maize, Zhenghong 505) and S intercropping; DM, dry matter; CP, crude protein; WSC, water-soluble carbohydrate; NDF, neutral detergent fiber; ADF, acid detergent fiber; RFV, relative feeding value; SEM, standard error of the mean. Means in the same column that do not share the same letters differ significantly (p < 0.05).

and

Table 4. Effects of maize varieties on the chemical composition of soybeans in maize–soybean intercropping.

Treatments	DM%	CP/%DM	WSC/%DM	NDF/%DM	ADF/%DM	RFV
M1S	27.04 c	21.04 b	6.90 b	44.00 c	29.10 b	140.15 b
M2S	18.63 d	23.79 a	3.81 d	49.10 a	32.14 a	121.10 c
M3S	32.01 a	18.27 d	8.55 a	30.49 d	21.50 c	220.29 a
M4S	30.42 b	19.35 c	5.73 c	46.75 b	32.58 a	126.43 c
SEM	2.58	1.04	0.86	3.61	2.22	20.02

M1S, M1 (maize, Rongyu Silage No. 1) and S (soybean, Nanxiadou 25) intercropping; M2S, M2 (maize, Yayu 04889) and S intercropping; M3S, M3 (maize, Demeiya No. 1) and S intercropping; M4S, M4 (maize, Zhenghong 505) and S intercropping; DM, dry matter; CP, crude protein; WSC, water-soluble carbohydrate; NDF, neutral detergent fiber; ADF, acid detergent fiber; RFV, relative feeding value; SEM, standard error of the mean. Means in the same column that do not share the same letters differ significantly (p < 0.05).

, maize varieties had significant effects on the chemical composition in maize–soybean intercropping (p < 0.05). According to Table 2, the DM content of M1 was the highest (36.86%), which was significantly higher than that of M2 and M3 (p < 0.05), followed by M4 (33.10%). The contents of CP (8.46% DM) and WSC (14.59% DM) in M3 were significantly higher than those in the other treatments (p < 0.05). The contents of NDF (55.55% DM) and ADF (35.00% DM) in M2 were the highest, and the differences were significant compared with the other treatments (p < 0.05). The RFV value of M2 was the lowest (103.26), and there was no significant difference among the other treatments.

According to Table 4, the DM content of soybeans in M3S was the highest (32.01%), while that of M2S was the lowest (18.63%). The CP content of soybeans in M2S was the highest (23.79% DM), followed by M1S (21.04% DM). The WSC content of soybeans in M3S was the highest (8.55% DM), followed by M1S and M2S. The contents of NDF and ADF in soybeans in M3S were significantly lower than those in the other treatments (p < 0.05), and the RFV value of soybeans in M3S was significantly higher than that in the other treatments (p < 0.05).

3.3. Effects of Maize Varieties on the Silage Quality of Maize–Soybean Intercropping

The maize varieties had significant effects on the silage quality of maize–soybean intercropping (p < 0.05). As shown in

Table 5. Effects of maize varieties on the silage quality of maize–soybean intercropping.

Treatments		DM%	CP /%DM	WSC /%DM	NDF /%DM	ADF /%DM	RFV	pH	NH3-N/TN /%
M	M1	33.71 a	7.08 d	7.16 b	43.00 d	26.04 d	148.44 a	4.22 a	3.15 a
	M2	22.19 c	7.83 b	2.79 d	52.13 a	33.01 a	112.76 d	3.96 c	4.29 a
	M3	20.92 d	9.12 a	8.08 a	48.81 b	29.10 b	126.28 c	4.18 ab	4.14 a
	M4	32.54 b	7.50 c	4.15 c	46.29 c	27.09 c	136.24 b	4.11 b	3.65 a
S	S1	27.12 c	20.81 a	1.30 b	42.31 c	31.55 b	141.47 b	5.45 b	13.98 b
	S2	15.85 d	17.44 b	0.75 c	47.83 a	36.06 a	118.36 c	6.62 a	37.56 a
	S3	30.79 a	20.63 a	3.59 a	26.62 d	18.95 c	259.41 a	5.15 c	5.46 c
	S4	28.30 b	19.02 b	1.08 bc	44.78 b	34.60 a	128.80 bc	5.50 b	14.78 b
MS	M1S1	24.53 b	11.67 c	4.00 b	39.13 c	25.15 c	164.79 a	4.22 bc	6.69 a
	M2S2	20.29 d	12.59 b	1.63 d	52.99 a	34.80 a	108.54 c	4.11 c	6.36 a
	M3S3	23.10 c	13.44 a	4.45 a	39.42 c	25.42 c	163.20 a	4.46 a	3.97 b
	M4S4	31.55 a	11.33 d	2.68 c	44.37 b	27.89 b	140.97 b	4.29 b	6.52 a
SEM		1.56	1.41	0.64	1.94	1.39	11.06	0.23	2.68

M, maize sole silage; S, soybean sole silage; MS, mixture silage of maize and soybean; M1, Rongyu Silage No. 1 (maize); M2, Yayu 04889 (maize); M3, Demeiya No. 1 (maize); M4, Zhenghong 505 (maize); S1, the soybeans intercropped with M1; S2, the soybeans intercropped with M2; S3, the soybeans intercropped with M3; S4, the soybeans intercropped with M4; DM, dry matter; CP, crude protein; WSC, water-soluble carbohydrate; NDF, neutral detergent fiber; ADF, acid detergent fiber; RFV, relative feeding value; NH₃-N, ammonia nitrogen; TN, total nitrogen; SEM, standard error of the mean. Means in the same column that do not share the same letters differ significantly (p < 0.05).

, compared with S and MS, M had the lowest CP content, the highest WSC content, and the best fermentation quality in each treatment. S was the opposite. The nutrient composition and fermentation quality of MS were between those of M and S. In M, the content of DM in M1 was the highest (33.71%), followed by M4. The contents of CP (9.12% DM) and WSC (8.08% DM) in M3 were significantly higher than those in the other treatments (p < 0.05). The contents of NDF (52.13% DM) and ADF (33.01% DM) in M2 were the highest, and the RFV value was the lowest (112.76%). The pH value of all treatments in M was less than 4.22, and the pH value of M2 was the lowest, while the NH₃-N/TN level was similar among all treatments in M. In S, the content of DM (30.79%) in S3 was significantly higher than that in other treatments (p < 0.05), and the contents of CP (20.63% DM) and WSC (3.95% DM) were the highest, the contents of NDF (26.62% DM) and ADF (18.95% DM) were the lowest, and the value of RFV (259.41%) was the highest. The pH value and NH₃-N/TN of S3 were the lowest, but the pH value of all treatments in soybean silage was greater than 5; the pH value (6.62) and NH₃-N/TN (37.56%) in S2 were the highest, which were significantly higher than those in the other treatments (p < 0.05). In MS, the content of DM in M4S4 was the highest (31.55%), followed by M1S1 and M2S.2. The contents of CP and WSC in M3S3 were significantly higher than those in the other treatments (p < 0.05). The contents of NDF (52.99% DM) and ADF (34.80% DM) in M2S2 were the highest, and the RFV value (108.54) was the lowest. The pH value of mixed silage was less than 4.5, and M2S2 had the lowest pH value (4.11). The NH₃-N/TN of M3S2 was the lowest (3.97%), and there were no significant differences among the other treatments.

3.4. Comprehensive Evaluation

The fuzzy mathematical membership function method was used to calculate the membership function values of the related indicators involved in the utilization of fresh feeding and silage for the four treatments in this study. According to

Table 6. Membership function analysis of forage utilized for fresh feeding.

Treatments		DMY	DM	WSC	CP	NDF	ADF	RFV	Average	Ranking
M1S	M	1.00	1.00	0.00	0.00	1.00	0.58	1.00	0.57	2
	S	0.99	0.63	0.48	0.50	0.27	0.31	0.19
M2S	M	0.00	0.01	0.27	0.53	0.00	0.00	0.00	0.13	4
	S	0.00	0.00	0.00	1.00	0.00	0.04	0.00
M3S	M	0.07	0.00	1.00	1.00	0.36	1.00	0.48	0.66	1
	S	0.28	1.00	1.00	0.00	1.00	1.00	1.00
M4S	M	0.76	0.74	0.04	0.15	0.65	0.76	0.58	0.44	3
	S	1.00	0.88	0.26	0.20	0.13	0.00	0.05

M1S, M1 (maize, Rongyu Silage No. 1) and S (soybean, Nanxiadou 25) intercropping; M2S, M2 (maize, Yayu 04889) and S intercropping; M3S, M3 (maize, Demeiya No. 1) and S intercropping; M4S, M4 (maize, Zhenghong 505) and S intercropping; M, membership function value of maize in the corresponding treatment; S, membership function value of soybeans in the corresponding treatment; Average, average value of membership function; Ranking, ranking of average value of membership function; DMY, dry mater yield; DM, dry matter; CP, crude protein; WSC, water-soluble carbohydrate; NDF, neutral detergent fiber; ADF, acid detergent fiber; RFV, relative feeding value.

and

Table 7. Membership function analysis of silage utilization.

Treatments		DM	WSC	CP	NDF	ADF	RFV	PH	NH3-N/TN	Average	Ranking
M1S	M	1.00	0.83	0.00	0.26	0.26	1.00	0.00	1.00	0.63	2
	S	0.75	0.19	1.00	1.00	1.00	0.16	0.79	0.73
	MS	0.38	0.84	0.16	1.00	1.00	1.00	0.68	0.00
M2S	M	0.10	0.00	0.37	0.00	0.00	0.00	1.00	0.00	0.15	4
	S	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
	MS	0.00	0.00	0.59	0.00	0.00	0.00	1.00	0.12
M3S	M	0.00	1.00	1.00	1.00	1.00	0.40	0.15	0.13	0.74	1
	S	1.00	1.00	0.94	0.36	0.56	1.00	1.00	1.00
	MS	0.25	1.00	1.00	0.98	0.97	0.97	0.00	1.00
M4S	M	0.91	0.26	0.20	0.14	0.09	0.70	0.42	0.57	0.48	3
	S	0.83	0.12	0.47	0.64	0.85	0.66	0.07	0.71
	MS	1.00	0.37	0.00	0.62	0.72	0.58	0.49	0.06

M1S, M1 (maize, Rongyu Silage No. 1) and S (soybean, Nanxiadou 25) intercropping; M2S, M2 (maize, Yayu 04889) and S intercropping; M3S, M3 (maize, Demeiya No. 1) and S intercropping; M4S, M4 (maize, Zhenghong 505) and S intercropping; M, membership function value of maize sole silage in the corresponding treatment; S, membership function value of soybean sole silage in the corresponding treatment; MS, membership function value of maize–soybean mixed silage in the corresponding treatment; Average, average value of membership function; Ranking, ranking of average value of membership function; DM, dry matter; CP, crude protein; WSC, water-soluble carbohydrate; NDF, neutral detergent fiber; ADF, acid detergent fiber; RFV, relative feeding value; NH₃-N, ammonia nitrogen; TN, total nitrogen.

, when it was used as fresh feed, M3S (0.66) > M1S (0.57) > M4S (0.44) > M2S (0.13), M3 had the highest membership function value. When used as silage, M3S (0.74) > M1S (0.63) > M4S (0.48) > M2S (0.15), M3S had the highest membership function value.

4. Discussion

4.1. Effects of Maize Varieties on Yield, Plant Height, and Chemical Composition in Maize–Soybean Intercropping

Intercropping has proven to be an effective way to increase crop yield and land equivalent ratio. In this study, the yields of different maize varieties intercropped with soybeans showed differences. It was previously reported that the yield was affected by crop varieties [25,26]. In this study, except for the maize varieties, other factors—such as field management and climate conditions—were consistent. We found that the yields of different maize varieties were different, which may also be related to maturity. In this study, there were no significant differences in biomass yield between the four maize varieties, but the dry matter yield of Rongyu Silage No. 1 and Zhenghong 505 was significantly higher than that of Yayu 04889 and Demeiya No. 1. This may be because Rongyu Silage No. 1 and Zhenghong 505 were in the late milk-ripening stage, while Yayu 04889 and Demeiya No. 1 were in the early milk-ripening stage, as it was previously reported that the more mature the maize, the more dry matter content [27]. In this study, the yield and plant height of soybeans in the different treatments were significantly different (p < 0.05); it may be the case that different maize varieties had different morphological characteristics and had different shading effects on soybeans [28], leading to the corresponding changes in soybean morphology, growth, and development, as well as corresponding changes in biomass yield and plant height [29].

Nutrient composition is the key index of forage quality, and can directly reflect the feeding value of forage. Javanmard et al. found that maize varieties have effects on nutrient composition in maize–bean intercropping [30]. We also found that the nutrient composition of maize may be affected by the growth period at harvest of the different varieties. In this study, the DM content of Rongyu Silage No. 1 and Zhenghong 505 was higher than that of Yayu 04889 and Demeiya No. 1. It was previously reported that the DM content increases with the increase in maturity [31]. The NDF and ADF contents of Yayu 04889 were the highest among the maize varieties, which may be because the increase in cellulose and lignin in maize stover at this stage was offset by the development of grain and the generation of a large number of carbon-containing compounds [32]. The CP and WSC contents of Demeiya No. 1 were the highest among the maize varieties, probably because in the late growth stage of maize carbohydrates are mainly generated and the nitrogen deposition rate decreases [33,34], which leads to a decrease in the proportion of CP in the maize. From the early dent stage to the blackline stage, starch accumulation leads to a sharp decrease in WSC [32]. For soybeans, those intercropped with Demeiya No. 1 had the highest contents of DM and WSC and the lowest contents of NDF and ADF, while soybeans intercropped with Yayu 04889 had the highest CP content. The DM content of soybeans in intercropped with Yayu 04889 and Nanxiadou 25 was the lowest, which may have been due to the influence of the shade environment of the maize and the shade-avoidance reaction of soybeans, represented by the increase in plant height, the decrease in stem diameter, and the increase in internode length, eventually leading to the decrease in dry matter accumulation [35]. In this study, the NDF and ADF contents of soybeans in the intercropping of Yayu 04889 and Nanxiadou 25 were the highest. This is because the intercropped soybeans were affected by maize shading, causing the soybeans to obtain more light by increasing their plant height [36,37], resulting in increased NDF and ADF contents. Studies have shown that the cellulose content of legumes is usually lower than that of grasses [38], which is consistent with the results of this study. Hartwig et al. [39] found that raffinose, stachyose, and sucrose in the WSC of soybeans were negatively correlated with the CP contents. In this study, CP was negatively correlated with the contents of WSC in the four treatments.

4.2. Effects of Maize Varieties on the Silage Quality of Maize–Soybean Intercropping

Silage is the main method to preserve forage nutrition. Previous research shows that different varieties of whole-plant maize silage have significant differences in nutritional components [40], indicating that variety selection has an important impact on the nutritional value of silage; this study shows similar results. In this study, the DM content of soybeans in the intercropping of Yayu 04889 and Nanxiadou 25 was the lowest before and after silage, and the pH value and NH₃-N/TN of this treatment were the highest among all of the treatments. The reason for this is that the DM content of the soybeans was affected by the shade effect of the maize [35], and DM loss and fermentation quality of silage are directly related to the DM content of the raw materials at harvest, where low DM content of soybeans may reflect poor stability of the silage fermentation process [41]. Compared with that before ensiling, the CP content of soybeans sole silage in the intercropping of Yayu 04889 and Nanxiadou 25 decreased by 6.35%, which may be because Clostridium becomes active under the conditions of low DM content and high pH value (>5), decomposing amino acids into NH₃ and other mixtures [42]. WSC is the main substrate used by lactic acid bacteria in silage [43]; the more abundant the WSC content, the faster the silage fermentation and the less nutrient loss. Compared with before ensilage, the DM content of maize sole silage in the intercropping of Yayu 04889 and Nanxiadou 25 was almost completely retained, because the treatment consumed a large amount of WSC for the propagation of lactic acid bacteria, which rapidly reduced the pH value, creating an acidic environment, and avoiding the proliferation of undesirable microorganisms and consumption of nutrients. Generally, the contents of NDF and ADF show almost no change before and after silage, but in this study the NDF and ADF contents after ensilage were lower than those before ensilage in maize sole silage, partly because maize is rich in hemicellulose, which is sensitive to low pH and partially hydrolyzed under acidic conditions [44]. In this study, the comprehensive quality of maize–soybean mixed silage was better than that of sole silage. A large number of studies have shown that, compared with maize sole silage, mixed silage can increase CP content [45] and reduce NDF and ADF contents [46]. Compared with soybean sole silage, this can make up for the shortcomings of low WSC content and high buffer energy of soybeans [47], improving the fermentation quality.

5. Conclusions

Different maize varieties had significant effects on the biomass yield, nutritional composition, and silage quality of maize–soybean intercropping, among which M1S had the highest total dry matter yield, while M3S had the highest contents of CP and WSC in maize and the lowest NDF and ADF contents in soybean. After silage, the CP content of M3S mixed silage was higher and the NDF and ADF contents were lower than those of maize sole silage, the WSC content was higher than that of soybean sole silage, and the pH value and NH₃-N/TN were between those of maize and soybean sole silages. The results of membership function analysis showed that M3S was the best in fresh feeding and silage utilization, followed by M1S. Overall, it is recommended to promote M3S—i.e., Demeiya No 1. and Nanxiadou 25—intercropping and silage production in high-altitude areas such as Ganzi Prefecture.