Effect of Mixed Lactic Acid Bacteria on Silage Quality and In Vitro Digestibility of 16 Oat Forage Varieties in Qinghai–Tibet Plateau

Abstract

The production of artificial planted forage is important for the development of animal husbandry in the Qinghai–-Tibet Plateau, and oat forage is one of the main artificial planted forages in the area. However, the suitable oat varieties for harvesting and preparing silage feed in this region are still unclear. To investigate suitable oat forage varieties for silage production and the potential feeding value of different oat forage varieties, 16 oat forage varieties planted in Qinghai Province were selected in this experiment. These oat forages were subjected to two treatments: a group with no inoculants (CK) and a group with self-selected lactic acid bacteria (LAB) inoculants (IN). After 90 days of ensiling, silage quality and in vitro digestibility of the 16 oat forage varieties were determined. The results showed that all oat forage varieties ferment well after ensiling (pH < 4), the CK group had a silage pH range of 3.62–3.95, and the IN group had a silage pH range of 3.68–3.83. Tianyan No.1, Qingtian No.2, and Tianyan No.3 were in the top three in RFV and RFQ rankings in the CK group, while Qinghai 444, Tianyan No.1, and Tianyan No.3 were in the lead in GI rankings. Tianyan No.1, Qingtian No.2, and Everleaf 126 were in the lead in RFV and RFQ rankings in the IN group, while Qinghai 444, Titan, and Tianyan No.1 were in the top three in GI rankings. The dry matter digestibility and 72 h cumulative gas production of the IN group were higher than that of the CK group (p < 0.05). Based on principal component analysis and membership function comprehensive evaluation, Tianyan No.1, Qinghai 444, and Tianyan No.3 ranked the top three, demonstrating that these three oat forage varieties are suitable for silage processing in the Qinghai–Tibet Plateau region.

1. Introduction

Oats (Avena sativa L.), as an annual forage crop, are characterised by their tender and succulent stems and leaves, rich nutritional content, and remarkable tolerance to drought, cold, and poor soil conditions. They exhibit extensive adaptability and high nutritional value, making them particularly valuable in high-altitude pastoral regions [1,2]. However, the traditional hay-making method has drawbacks, such as serious nutritional loss and limitation by geographical and climatic factors [3,4], which makes it difficult to meet the sustained demand for high-quality forage in high-altitude pastoral regions. Silage technology, which preserves fresh forage through microbial fermentation, can prolong storage time, reduce nutrient loss, and provide balanced nutrition and has become an effective way to solve the forage problem in high-altitude pastoral areas [5]. However, the raw material characteristics of different oat forage varieties vary greatly; we hypothesise that these varietal differences will lead to significant variations in silage quality parameters, including fermentation characteristics, nutrient preservation, and fibre degradation.

It is noteworthy that unfavourable conditions such as low temperature in high-altitude regions often lead to unstable silage fermentation, which affects silage quality. To improve silage quality, we further propose that high-quality lactic acid bacteria (LAB) can be added during the ensiling process to promote the production of organic acids and inhibit the reproduction of harmful microorganisms, thus enhancing the stability and nutritional value of silage [6]. In assessing silage quality, the in vitro gas production method has unique advantages as a test method that simulates rumen digestion in ruminants [7]. This method not only accurately evaluates the digestibility of organic matter in the rumen but also determines the effects of feed inoculants, which provides a scientific basis for the comprehensive evaluation of silage quality [8]. The in vitro gas production method allows for the assessment of the nutritional value and digestibility of silage from different oat forage varieties, and the effect of inoculants such as LAB on silage fermentation and digestibility is also investigated.

In this study, 16 oat forage varieties from the Qinghai–Tibet Plateau region were selected to evaluate their silage conventional nutrients and in vitro digestibility of their silage. By combining the in vitro gas production method with traditional evaluation methods, we comprehensively assessed silage quality and identified oat forage varieties suitable for the Qinghai–Tibet Plateau region. These findings provide important theoretical support and practical guidance for the development of forage resources, optimisation of silage preparation, and sustainable development of animal husbandry in high-altitude pastoral regions.

2. Materials and Methods

2.1. Materials

The 16 oat forage varieties selected for the experiment were Qingtian No.2, Baiyan No.7, Baler, Haymaker, Hanma, Bayan No.6, Everleaf 126, Titan, Youmu No.1, Sweet oat, Magnum, Forage plus, Qinghai 444, Tianyan No.3, Monica, and Tianyan No.1, which were all provided by Qinghai Academy of Animal Science and Veterinary Medicine. The experiment was arranged in a completely randomised block design with three replicates per variety. Each plot measured 15 m² (3 m × 5 m) and consisted of 16 rows, with 0.5 m spacing between adjacent plots. The trial was conducted in Taxiu Township, Guinan County, Qinghai Province (100.75 °E, 35.57 °N), at an altitude of 3266 m, with sowing performed on 21 May 2021. All varieties were harvested at the milk stage with a stubble height of 10 cm. The silage was prepared using self-developed LAB inoculants named “Silage Mate” by the research group, including Lactobacillus plantarum 160, Pediococcus pentosaceus 260, and Lactobacillus buchneri 225 (viable bacterial count ≥ 2.0 × 10¹⁰ CFU/g). After harvesting, the oat forages were immediately chopped to 1–2 cm pieces using a forage cutter and thoroughly mixed. A portion of the mixed material was compacted and vacuum-sealed in polyethylene bags (260 mm × 350 mm) using a vacuum sealer (DZQ-390, Anshengke, Fujian, China), with each weighing about 1 kg. The remaining portion was uniformly sprayed with the inoculant, packed into polyethylene bags, each weighing about 1 kg, and vacuum-sealed using the same equipment. Three replicates were prepared for each treatment, and they were left to ferment in the local outdoor environment where the daily temperature ranged between 25 °C and 32 °C. After 90 days of ensiling, the bags were opened, and about 1000 g of silage samples were dried in an oven until constant weight. The dried samples were then crushed in a shredder (2500C, Hongtaiyang, Yongkang, China) and used for the determination of nutritional quality and in vitro fermentation parameters.

2.2. Determination of Conventional Nutrients

The silage samples of each variety were dried in an oven at 65 °C for 48 h to constant weight to determine the dry matter (DM) content [9]. Crude protein (CP) was determined using the Kjeldahl method [10]. Water-soluble carbohydrate (WSC) was determined using the anthrone-sulfuric acid colourimetric method, with a plant water-soluble carbohydrate assay kit produced by Suzhou Keming Biotechnology Co., Ltd.(Suzhou, China) Neutral detergent fibre (NDF) and acid detergent fibre (ADF) were determined by the Van Soest detergent fibre method [11]. Crude ash (Ash) was determined by burning in a muffle furnace.

2.3. Determination of Silage Quality

For each raw material and treated silage sample, 20 g was weighed and mixed with 180 mL of distilled water. The mixtures were sealed and stored 4 °C overnight to allow for aqueous extraction. The next day, the mixtures were filtered through 4 layers of gauze, and a portion of the filtrate was used to determine the pH using a pH meter (pHS-10). The filtrate was further filtered through a 0.22 μm membrane into a sample vial. The concentrations of lactic acid (LA), acetic acid (AA), propionic acid (PA), and butyric acid (BA) were determined using an ultra-performance liquid chromatography system (Thermo Scientific UltiMate 3000, Thermo Fisher Scientific, MA, USA) [12], with the chromatography conditions of an RSpak KC-811 column, mobile phase 0.1% H₃PO₄, flow rate 0.5 mL·min⁻¹, and column temperature 55 °C.

2.4. Feed Quality Rating Index

The feed quality rating index includes relative feeding value (RFV), relative forage quality (RFQ), and forage grading index (GI).

2.4.1. Relative Feeding Value

Relative feeding value (RFV) is calculated using the formula as follows [13]:

$$\mathrm{R}\mathrm{F}\mathrm{V}=\mathrm{D}\mathrm{M}\mathrm{I}\times \mathrm{D}\mathrm{D}\mathrm{M}/1.29,$$

$$\mathrm{D}\mathrm{M}\mathrm{I}=120/\mathrm{N}\mathrm{D}\mathrm{F},$$

$$\mathrm{D}\mathrm{D}\mathrm{M}=88.9-0.779\times \mathrm{A}\mathrm{D}\mathrm{F},$$

where DMI is dry matter intake (%BW, body weight); NDF is neutral detergent fibre content of forage (g/kgDM); DDM is digestible dry matter (g/kgDM); and ADF is acid detergent fibre content of forage (g/kgDM).

2.4.2. Relative Forage Quality

The formula for relative forage quality (RFQ) is as follows [13]:

$$\mathrm{R}\mathrm{F}\mathrm{Q}=1.9499\mathrm{R}\mathrm{F}\mathrm{V}-67.038({\mathrm{R}}^2=0.7552).$$

2.4.3. Forage Grading Index

The forage grading index (GI) formula is calculated as follows [14]:

$$\mathrm{G}\mathrm{I}(\mathrm{M}\mathrm{J}/\mathrm{d})={\mathrm{N}\mathrm{E}}_{\mathrm{L}}\times \mathrm{V}\mathrm{D}\mathrm{M}\mathrm{I}\times \mathrm{C}\mathrm{P}/\mathrm{N}\mathrm{D}\mathrm{F},$$

In calculating GI, NDF was used to predict VDMI, and ADF was used to predict NE_L for forage:

$$\mathrm{V}\mathrm{D}\mathrm{M}\mathrm{I}=1.2/\mathrm{N}\mathrm{D}\mathrm{F}\times \mathrm{B}\mathrm{W},$$

$${\mathrm{N}\mathrm{E}}_{\mathrm{L}}=[1.085-(0.0124\times \mathrm{A}\mathrm{D}\mathrm{F}\left)\right]\times 9.29,$$

where NE_L is net energy for lactation (MJ/kg); VDMI is voluntary dry matter intake (kg/d); CP is crude protein content of forage (g/kgDM); NDF is neutral detergent fibre content of forage (g/kgDM); ADF is the acid detergent fibre content of forage (g/kgDM); and BW is the body weight of the cow (set at 600 kg).

2.5. Determination of In Vitro Digestibility

2.5.1. Collection of Rumen Fluid

The rumen fluid was obtained from four grazing male yaks (average body weight: 600 kg) of close weight from Qingbaijiang slaughterhouse in Chengdu, Sichuan Province, China. After extracting the rumen contents from the animals, they were immediately placed into pre-warmed thermos bottles pre-filled with CO₂ gas and quickly transported back to the laboratory for freezing and storage at −20 °C. At the time of use, they were thawed in a water bath at 39 °C, filtered through four layers of cheesecloth into a receiving bottle, and water-bathed at 39 °C with continuous ventilation of CO₂ to ensure anaerobic conditions.

2.5.2. Preparation of Artificial Buffer Solution

The artificial rumen buffer was prepared according to the method of Goering et al. [15]; 520.2 mL distilled water + 0.1 mL of solution A (13.2 g CaCl₂·2H₂O, 10.0 g MnCl₂·4H₂O, 1.0 g CoCl₂·6H₂O, 8.0 g FeCl₃·6H₂O to 100 mL with distilled water) + 208.1 mL of solution B (4 g NH₄HCO₃, 35 g NaHCO₃ to 1000 mL with distilled water) + 208.1 mL of solution C (5.7 g Na₂HPO₄, 6.2 g KH₂PO₄, 0.6 g MgSO₄·7H₂O to 1000 mL with distilled water) + 1.0 mL of solution D (100 mg of resazurin dissolved in 100 mL distilled water) was mixed in a 2000 mL conical flask. Before use, 62.5 mL of solution E (625 mg cysteine hydrochloride dissolved in 95 mL distilled water, followed by 4 mL of 1 mol/L NaOH and 625 mg Na₂S·9H₂O) was added and mixed well, and the reaction system was placed in a 39 °C constant temperature water bath. During this process, CO₂ was continuously injected into the system until the buffer was nearly colourless.

2.5.3. Determination of Gas Production In Vitro

The ANKOM RFS in vitro gas production system and a computer equipped with the corresponding software were used for the measurements. An amount of 1 g of fermentation substrate was accurately weighed in a 250 mL gas-producing bottle. The bottle was preheated in a constant-temperature air-bath shaker for 60 min at 39 °C. Filtered rumen fluid was mixed well with pre-prepared artificial rumen buffer solution at a 1:4 volume ratio, ensuring that CO₂ was continuously passed through to maintain an anaerobic environment. A 39 °C water bath was continuously maintained during this process. An amount of 150 mL of well-mixed rumen fluid were added to the gas-producing bottle, and the bottle was tightly sealed. The gas-producing bottle with the cap screwed on was placed in constant-temperature air-bath shaker and incubated at 39 °C for 72 h [16].

The software automatically recorded the cumulative and absolute pressures in the gas-producing bottle at each time point on the computer connected to the in vitro gas-producing system. The time points for this experiment were 2 h, 4 h, 8 h, 12 h, 24 h, 36 h, 48 h, and 72 h.

According to the ANKOM RFS system manual, the formula for converting gas production pressure (psi) to gas production volume (mL) at 39 °C is as follows:

$${\mathrm{V}}_{\mathrm{x}}={\mathrm{V}}_{\mathrm{j}}\times {\mathrm{P}}_{\mathrm{p}\mathrm{s}\mathrm{i}}\times 0.068004084,$$

where V_x is the volume of gas production at 39 °C (mL); V_j is the headspace volume of the gas production bottle (mL); and P_psi is the cumulative pressure at a point in time recorded by the software (psi).

2.5.4. Determination of In Vitro Fermentation Parameters

At the end of incubation, the fermentation was terminated in an ice-water bath, and the pH was determined using a pH meter (pHS-10). Ammonia nitrogen (NH₃-N) was determined by the improved colorimetric method described by Feng Zongci et al. [17].

Nylon cloths were prepared, numbered, dried, and weighed. The corresponding culture solution was filtered through the nylon cloth, and the filtrate was dispensed into 15 mL tubes and stored at −20 °C for the determination of NH₃-N. We rinsed the nylon cloth and sample residue with distilled water, placed the washed nylon cloth and residue together into an oven at 65 °C to dry for 48 h, and then transferred them into a desiccator to cool down and weighed them.

The dry matter degradation rate was calculated using the following formula:

$$\mathrm{D}\mathrm{M}\mathrm{D}\left(\mathrm{\%}\mathrm{D}\mathrm{M}\right)=\left[\right(\mathrm{M}+{\mathrm{M}}_1-{\mathrm{M}}_2)/\mathrm{M}]\times 100,$$

where DMD is the dry matter degradability; M is the weight of dry matter of the sample (g); M₁ is the weight of the nylon cloth (g); and M₂ is the total weight of the dry matter of the fermentation residue and the nylon cloth (g).

2.6. Statistical Analysis

The data were initially collated using Excel 2019.0, followed by statistical analysis using SPSS 25.0 software. A one-way ANOVA model was employed to investigate the measured indicators of each oat forage variety. The general linear model was applied to analyse the effects of variety, treatment, and their interactions on silage evaluation indicators, with multiple comparisons conducted using Duncan’s method. Correlation analysis and principal component analysis were performed on the measured relevant indicators using SPSS to extract the principal components. The membership function was used to calculate the membership values of the principal component scores for each variety, and comprehensive scores were computed based on their respective weights [18,19].

The membership function was calculated as follows:

$${\mu }_{\mathrm{i}\mathrm{n}}=({\mathrm{X}}_{\mathrm{i}\mathrm{n}}-{\mathrm{X}}_{\mathrm{i}\mathrm{m}\mathrm{i}\mathrm{n}})/({\mathrm{X}}_{\mathrm{i}\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{X}}_{\mathrm{i}\mathrm{m}\mathrm{i}\mathrm{n}}),$$

where μ_in is the membership value of the ith principal component score of the nth variety; X_in is the score of the ith principal component of the nth variety; X_imin is the minimum score of the ith principal component of all varieties; and X_imax is the maximum score of the ith principal component of all varieties.

The weights of the membership values were calculated using the following formula:

$${\mathrm{w}}_{\mathrm{i}}={\lambda }_{\mathrm{i}}/{\Sigma }_{\mathrm{i}=1}^{\mathrm{P}}{\lambda }_{\mathrm{i}},$$

where w_i is the weight of the ith principal component; λ_i is the contribution rate of the ith principal component; and p is the number of extracted principal components.

The composite score was calculated using the following formula:

$${\mathrm{D}}_{\mathrm{n}}={\Sigma }_{\mathrm{i}=1}^{\mathrm{p}}\left({\mathrm{w}}_{\mathrm{i}}\times {\mu }_{\mathrm{in}}\right),$$

where D_n is the composite score of the nth variety; w_i is the weight of the ith principal component; μ_in is the membership value of the ith principal component score of the nth variety; and p is the number of extracted principal components.

3. Results

3.1. Silage Nutrient Composition of Different Oat Forage Varieties

By analysing the silage nutrient composition of different oat forage varieties, we observed significant variations in silage nutrient composition among different oat forage silage; the results are shown in

Table 1. Effect of lactic acid bacterial inoculants on silage nutritional quality of 16 oat forage varieties.

Variety	Treatment	CP (g/kgDM)	Ash (g/kgDM)	WSC (g/kgDM)	NDF (g/kgDM)	ADF (g/kgDM)
Qingtian No.2	CK	4.4 ± 0.07 g	4.64 ± 0.05 ghij	12.15 ± 0.18 a	47.88 ± 1.3 i	22.29 ± 0.6 f
	IN	4.49 ± 0.23 hi	4.73 ± 0.02 h	12.41 ± 0.63 a	47.2 ± 0.27 g	19.37 ± 0.24 g
Baiyan No.7	CK	4.65 ± 0.05 f	4.72 ± 0.03 efgh	8.22 ± 0.45 b	53.06 ± 0.94 defg	27.16 ± 0 bc
	IN	4.74 ± 0.07 f	4.87 ± 0.07 g	8.55 ± 0.34 b	52.05 ± 2 bcd	21.67 ± 0.38 def
Baler	CK	4.56 ± 0.05 f	5.72 ± 0.01 b	5.65 ± 0.4 de	54 ± 0.45 d	25.42 ± 0.22 cd
	IN	4.69 ± 0.08 fg	5.85 ± 0.03 b	5.82 ± 0.08 efg	53.69 ± 0.02 b	21.19 ± 0.23 ef
Haymaker	CK	4.84 ± 0.06 e	4.68 ± 0.09 fghi	7.82 ± 0.36 bc	53.75 ± 0.29 de	25.6 ± 0.35 bcd
	IN	4.91 ± 0.09 e	4.75 ± 0.04 h	8.19 ± 0.68 b	52.18 ± 0.55 bc	23.64 ± 0.75 c
Hanma	CK	4.44 ± 0.06 g	4.89 ± 0.08 de	6.22 ± 0.34 d	52.21 ± 0.48 g	23.89 ± 0.8 def
	IN	4.55 ± 0.05 gh	4.96 ± 0.03 fg	6.3 ± 0.29 de	51.13 ± 0.28 cd	22.29 ± 0.68 d
Bayan No.6	CK	5.43 ± 0.03 c	4.85 ± 0.08 def	3.99 ± 0.16 h	59.12 ± 0.56 a	31.03 ± 1.84 a
	IN	5.5 ± 0.08 c	5.02 ± 0.09 f	4.02 ± 0.07 h	53.33 ± 0.69 b	23.5 ± 0.73 c
Everleaf 126	CK	4.41 ± 0.06 g	4.8 ± 0.03 defg	3.1 ± 0.13 i	53.77 ± 0.42 de	26.87 ± 1.84 bc
	IN	4.5 ± 0.07 hi	4.96 ± 0.02 fg	3.22 ± 0.58 i	48.06 ± 0.95 fg	20.9 ± 0.4 f
Titan	CK	6.15 ± 0.09 a	4.45 ± 0.34 j	4.86 ± 0.37 fg	53.52 ± 0.03 def	30.41 ± 2.58 a
	IN	6.19 ± 0.05 a	4.49 ± 0.05 j	5.16 ± 0.18 g	50.43 ± 1.2 de	21.97 ± 0.41 de
Youmu No.1	CK	4.64 ± 0.02 f	5.22 ± 0.06 c	5.57 ± 0.18 de	55.33 ± 0.67 c	29.2 ± 0.82 a
	IN	4.78 ± 0.15 ef	5.4 ± 0.07 d	6.01 ± 0.16 ef	52.29 ± 0.34 bc	28.48 ± 0.39 a
Sweet Oat	CK	5.47 ± 0.05 c	4.96 ± 0.02 d	4.59 ± 0.69 gh	52.55 ± 0.22 efg	27.43 ± 0.47 b
	IN	5.58 ± 0.05 c	5.16 ± 0.05 e	5.24 ± 0.33 fg	51.58 ± 0.98 cd	26.43 ± 1.35 b
Magnum	CK	5.81 ± 0.06 b	4.6 ± 0.09 hij	7.17 ± 0.45 c	57.12 ± 1.44 b	24.8 ± 0.47 de
	IN	5.92 ± 0.04 b	4.66 ± 0.03 hi	7.22 ± 0.47 c	56.76 ± 0.17 a	23.29 ± 0.39 c
Forage plus	CK	4.56 ± 0.08 f	4.64 ± 0.02 ghij	4.09 ± 0.4 h	52.44 ± 0.44 fg	24.03 ± 0.95 def
	IN	4.69 ± 0.04 fg	4.74 ± 0.01 h	4.35 ± 0.43 h	51.45 ± 1.05 cd	22.21 ± 0.4 de
Qinghai 444	CK	6.2 ± 0.09 a	6.46 ± 0.05 a	7.22 ± 0.3 c	50.53 ± 0.16 h	24.67 ± 0.8 de
	IN	6.28 ± 0.02 a	6.5 ± 0.04 a	7.33 ± 0.22 c	49.08 ± 0.98 ef	23.27 ± 0.35 c
Tianyan No.3	CK	5.04 ± 0.04 d	4.49 ± 0.05 ij	6.22 ± 0.25 d	48.76 ± 0.11 i	23.41 ± 0.1 ef
	IN	5.11 ± 0.03 d	4.61 ± 0.08 i	6.87 ± 0.89 cd	47.45 ± 0.13 g	22.23 ± 0.46 d
Monica	CK	5.83 ± 0.03 b	5.37 ± 0.04 c	5.25 ± 0.55 ef	52.8 ± 0.16 defg	29.96 ± 0.62 a
	IN	5.91 ± 0.05 b	5.54 ± 0.04 c	5.62 ± 0.34 efg	51.54 ± 0.27 cd	28.81 ± 0.2 a
Tianyan No.1	CK	4.3 ± 0.04 h	4.2 ± 0.08 k	5.79 ± 0.29 de	44.31 ± 0.22 j	22.55 ± 0.33 f
	IN	4.39 ± 0.09 i	4.49 ± 0.17 j	6.12 ± 0.39 de	43.55 ± 0.39 h	21.86 ± 0.44 def
SEM		0.066	0.055	0.217	0.362	0.318
Inoculants		**	**	**	**	**
Variety		**	**	**	**	**
Inoculants × Variety		NS	NS	NS	**	**

Note: The data in the table are mean ± standard error. Different lowercase letters in the same column indicate significant differences between the same indicators of different oat forage varieties. **: p < 0.01; NS: not significant; CK: control group (silage without lactic acid bacteria added); IN: bacteria added group (silage with lactic acid bacteria added); DM: dry matter; SEM: standard error; CP: crude protein; WSC: water-soluble carbohydrate; NDF: neutral detergent fibre; ADF: acid detergent fibre; Ash: crude ash.

. It was found that both the variety and the LAB inoculants had highly significant effects on the CP, Ash, WSC, NDF, and ADF contents (p < 0.01); the interaction between varieties and LAB inoculants had a highly significant effect on the contents of NDF and ADF (p < 0.01). Compared with the control group CK, the addition of LAB in the IN group resulted in varying degrees of increase in the CP, Ash, and WSC contents across different oat forage varieties, while the contents of NDF and ADF decreased to varying extents.

In the CK group, the CP content of Qinghai 444 was significantly higher than that of other varieties except Titan (p < 0.01), and the CP content of Tianyan No.1 was significantly lower than that of other varieties (p < 0.01); the Ash content of Qinghai 444 was highly significantly higher than that of other varieties (p < 0.01), and that of Tianyan No.1 was significantly lower than that of other varieties (p < 0.01). The WSC content of Qingtian No.2 was significantly higher than other varieties (p < 0.01), and the WSC content of Everleaf 126 was significantly lower than other varieties (p < 0.01). The NDF content of Bayan No.6 was significantly higher than other varieties (p < 0.01), and the NDF content of Tianyan No.1 was significantly lower than other varieties (p < 0.01). The ADF content of Bayan No.6 was significantly higher than other varieties other than Titan, Monica, and Youmu No.1 (p < 0.01), and the ADF content of Qingtian No.2 was significantly lower than other varieties other than Qinghai 444, Forage plus, Hanma, Tianyan No.3, and Tianyan No.1 (p < 0.01).

In IN group, the CP content of Qinghai 444 was significantly higher than that of other varieties except Titan (p < 0.01), and the CP content of Tianyan No.1 was significantly lower than that of other varieties except Qingtian No.2 and Everleaf 126 (p < 0.01); the Ash content of Qinghai 444 was significantly higher than that of other varieties (p < 0.01), and that of Titan was significantly lower than that of Tianyan No.1 (p < 0.01). The WSC content of Qingtian No.2 was significantly higher than that of other varieties (p < 0.01), and that of Everleaf 126 was significantly lower than that of other varieties (p < 0.01). The NDF content of Magnum was significantly higher than that of other varieties (p < 0.01), and that of Tianyan No.1 was significantly lower than that of other varieties (p < 0.01); the ADF content of Monica was significantly higher than that of other varieties except Youmu No.1 (p < 0.01), and the ADF content of Qingtian 2 was significantly lower than that of other varieties (p < 0.01).

3.2. Silage Quality of Different Oat Forage Varieties

By analysing the silage quality of different oat forage varieties, as shown in

Table 2. Effect of lactic acid bacterial inoculants on silage quality of 16 oat forage varieties.

Variety	Treatment	pH	Lactic Acid (mg/g DM)	Acetic Acid (mg/g DM)	Propionic Acid (mg/g DM)	Butyric Acid (mg/g DM)
Qingtian No.2	CK	3.86 ± 0.01 d	32.35 ± 5.88 cde	2.37 ± 0.3 cde	ND	ND
	IN	3.69 ± 0.02 gh	46.41 ± 12.89 abc	1.74 ± 0.36 bcde	ND	ND
Baiyan No.7	CK	3.89 ± 0.02 bcd	35.09 ± 4.14 bcde	2.27 ± 0.32 cde	ND	ND
	IN	3.73 ± 0.01 ef	41.25 ± 2.02 abc	1.63 ± 0.21 cde	ND	ND
Baler	CK	3.89 ± 0.01 bcd	29.82 ± 3.01 de	1.97 ± 0.53 de	ND	ND
	IN	3.74 ± 0.03 de	40.14 ± 6.51 bc	1.25 ± 0.22 e	ND	ND
Haymaker	CK	3.91 ± 0.02 b	27.38 ± 3.68 e	2.3 ± 0.45 cde	ND	ND
	IN	3.74 ± 0.03 cde	41.1 ± 6.19 abc	1.25 ± 0.09 e	ND	ND
Hanma	CK	3.87 ± 0.01 cd	26.87 ± 7.14 e	3.03 ± 0.13 ab	ND	ND
	IN	3.69 ± 0.01 gh	43.76 ± 1.49 abc	1.56 ± 0.34 cde	ND	ND
Bayan No.6	CK	3.77 ± 0.01 e	46.37 ± 9.62 ab	1.88 ± 0.36 e	ND	ND
	IN	3.75 ± 0.02 cde	51.53 ± 4.04 a	1.76 ± 0.41 bcde	ND	ND
Everleaf 126	CK	3.66 ± 0.06 h	31.61 ± 8.52 cde	3.27 ± 0.32 a	ND	ND
	IN	3.68 ± 0 h	43.56 ± 1.74 abc	2.4 ± 0.6 ab	ND	ND
Titan	CK	3.95 ± 0.01 a	42.01 ± 3.65 abcd	2.61 ± 0.6 bcd	ND	ND
	IN	3.78 ± 0.02 bc	49.32 ± 3.69 ab	1.57 ± 0.11 cde	ND	ND
Youmu No.1	CK	3.89 ± 0 bcd	35.26 ± 4.35 bcde	1.9 ± 0.28 e	ND	ND
	IN	3.72 ± 0.01 efg	45.85 ± 6.62 abc	1.47 ± 0.19 de	ND	ND
Sweet Oat	CK	3.9 ± 0.01 bc	28.97 ± 12.73 e	2.9 ± 0.19 abc	ND	ND
	IN	3.79 ± 0.02 b	38.06 ± 3.29 c	1.7 ± 0.87 cde	ND	ND
Magnum	CK	3.76 ± 0.02 e	37.03 ± 9.99 abcde	2.33 ± 0.13 cde	ND	ND
	IN	3.77 ± 0.02 bcd	41.08 ± 5.32 abc	1.72 ± 0.46 bcde	ND	ND
Forage plus	CK	3.71 ± 0.01 fg	43.71 ± 4.07 abc	2.66 ± 0.09 abc	ND	ND
	IN	3.73 ± 0.01 ef	44.34 ± 3.42 abc	2.19 ± 0.3 abc	ND	ND
Qinghai 444	CK	3.75 ± 0.03 ef	31.61 ± 0.27 cde	1.78 ± 0.14 e	ND	ND
	IN	3.83 ± 0.04 a	39.98 ± 1.51 bc	1.39 ± 0.11 e	ND	ND
Tianyan No.3	CK	3.69 ± 0.02 gh	42.27 ± 5.75 abcd	2.83 ± 0.37 abc	ND	ND
	IN	3.68 ± 0.01 h	48.4 ± 4.04 abc	2.13 ± 0.25 abcd	ND	ND
Monica	CK	3.74 ± 0.03 ef	37.82 ± 6.31 abcde	1.91 ± 0.09 e	ND	ND
	IN	3.79 ± 0.02 b	43.25 ± 3.04 abc	1.49 ± 0.14 cde	ND	ND
Tianyan No.1	CK	3.62 ± 0.01 i	48.21 ± 1.77 a	3.08 ± 0.54 ab	ND	ND
	IN	3.7 ± 0.03 fgh	46.6 ± 8.58 abc	2.66 ± 0.23 a	ND	ND
SEM		0.009	0.849	0.064	ND	ND
Inoculants		**	**	**	ND	ND
Variety		**	**	**	ND	ND
Inoculants × Variety		**	NS	NS	ND	ND

Note: The data in the table are mean ± standard error. Different lowercase letters in the same column indicate significant differences between the same indicators of different oat forage varieties. **: p < 0.01; NS: not significant; CK: control group (silage without lactic acid bacteria added); IN: bacteria added group (silage with lactic acid bacteria added); DM: dry matter; SEM: standard error.

, variety and LAB inoculant had highly significant effects on pH, LA, and AA (p < 0.01); the interaction of both variety and LAB inoculant had highly significant effects on pH (p < 0.01), and no significant effects on LA and AA contents (p < 0.05). PA and BA contents were not detected in all varieties of oat forage silage. Regarding an IN comparison with CK, there was an overall increasing trend in the LA content, while there was an overall decreasing trend in the pH and AA content.

The pH of Qinghai 444 in the CK group was significantly higher than other varieties (p < 0.01), and the pH of Tianyan No.3 was significantly lower than other varieties other than Everleaf 126, Hanma, Qingtian No.2, and Tianyan No.1 (p < 0.01). The LA content ranged from 26.87 g/mg to 48.21 g/mg, and the AA content ranged from 1.78 g/mg to 3.27 g/mg. The pH was significantly higher in the IN group Titan than other varieties (p < 0.01) and significantly lower in Tianyan No.1 than other varieties (p < 0.01). The LA content ranged from 38.06 g/mg to 51.53 g/mg, and the AA content ranged from 1.25 g/mg to 2.66 g/mg.

3.3. Feed Quality Rating Index of Different Oat Forage Varieties

Through the analysis of the feed quality evaluation indices of various oat forage varieties, the results showed that the differences between RFQ and RFV was basically similar among the different varieties, and the specific data are shown in

Table 3. Effect of lactic acid bacterial inoculants on silage RFV, RFQ, and GI of 16 oat forage varieties.

Variety	Treatment	RFV	RFQ	GI (MJ/d)
Qingtian No.2	CK	139.03 ± 3.45 b	204.06 ± 6.74 b	10.38 ± 0.37 abc
	IN	145.48 ± 0.65 b	216.63 ± 1.27 b	11.4 ± 0.63 d
Baiyan No.7	CK	120.47 ± 0 f	167.86 ± 0 f	8.59 ± 0 bcde
	IN	128.83 ± 4.95 ef	184.18 ± 9.66 ef	9.59 ± 0.86 fgh
Baler	CK	119.04 ± 1.28 f	165.08 ± 2.5 f	8.06 ± 0.2 cde
	IN	125.42 ± 0.38 fg	177.52 ± 0.74 fg	8.87 ± 0.03 hi
Haymaker	CK	119.35 ± 1.08 f	165.69 ± 2.11 f	8.61 ± 0.04 bcde
	IN	125.66 ± 1.68 fg	178 ± 3.28 fg	9.56 ± 0.37 fgh
Hanma	CK	125.22 ± 0.8 de	177.14 ± 1.57 de	8.59 ± 0.2 bcde
	IN	129.62 ± 1.1 ef	185.71 ± 2.16 ef	9.37 ± 0.29 gh
Bayan No.6	CK	101.87 ± 3.22 j	131.59 ± 6.27 j	7.27 ± 0.34 de
	IN	123.55 ± 0.59 gh	173.87 ± 1.15 gh	10.37 ± 0 ef
Everleaf 126	CK	117.59 ± 2.79 fg	162.26 ± 5.43 fg	7.67 ± 0.18 cde
	IN	140.59 ± 2.88 c	207.09 ± 5.6 c	10.75 ± 0.27 de
Titan	CK	111.88 ± 3.46 i	151.11 ± 6.74 i	9.96 ± 0.25 abcd
	IN	132.47 ± 3 de	191.28 ± 5.85 de	13.25 ± 0.66 ab
Youmu No.1	CK	111.24 ± 2.39 i	149.87 ± 4.67 i	7.34 ± 0.26 de
	IN	118.69 ± 0.44 i	164.39 ± 0.85 i	8.55 ± 0.21 i
Sweet Oat	CK	119.54 ± 0.84 f	166.05 ± 1.64 f	9.87 ± 0.21 abcd
	IN	123.22 ± 3.19 gh	173.22 ± 6.22 gh	10.63 ± 0.57 de
Magnum	CK	113.36 ± 2.27 hi	153.99 ± 4.43 hi	6.32 ± 5.48 e
	IN	115.97 ± 0.19 i	159.09 ± 0.37 i	9.79 ± 0.09 fg
Forage plus	CK	124.5 ± 0.48 e	175.72 ± 0.94 e	8.72 ± 0.18 bcde
	IN	129.49 ± 2.69 ef	185.46 ± 5.26 ef	9.59 ± 0.47 fgh
Qinghai 444	CK	128.28 ± 1.26 d	183.09 ± 2.45 d	12.65 ± 0.36 a
	IN	134.15 ± 2.17 d	194.55 ± 4.24 d	13.89 ± 0.44 a
Tianyan No.3	CK	134.8 ± 0.35 c	195.81 ± 0.68 c	11.28 ± 0.04 ab
	IN	140.32 ± 0.92 c	206.57 ± 1.8 c	12.29 ± 0.09 c
Monica	CK	115.51 ± 0.5 gh	158.19 ± 0.97 gh	9.97 ± 0.09 abcd
	IN	119.95 ± 0.7 hi	166.86 ± 1.35 hi	10.83 ± 0.18 de
Tianyan No.1	CK	149.77 ± 0.27 a	224.99 ± 0.52 a	11.79 ± 0.1 a
	IN	153.53 ± 1.28 a	232.33 ± 2.5 a	12.6 ± 0.39 bc
SEM		1.244	2.425	0.213
Inoculants		**	**	**
Variety		**	**	**
Inoculants × Variety		**	**	NS

Note: The data in the table are mean ± standard error. Different lowercase letters in the same column indicate significant differences between the same indicators of different oat forage varieties. **: p < 0.01; NS: not significant; CK: control group (silage without lactic acid bacteria added); IN: bacteria added group (silage with lactic acid bacteria added); RFV: relative feeding value; RFQ: relative forage quality; GI: forage grading index.

. The variety and LAB inoculant had a highly significant effect on RFV, RFQ, and GI (p < 0.01); the interaction between the variety and LAB inoculant had a significant effect on RFV and RFQ (p < 0.01), and the difference on GI was not significant (p > 0.05). Compared to the CK group, the IN group showed varying degrees of increase in RFV, RFQ, and GI across different oat forage varieties.

The three varieties with the highest RFV and RFQ in the CK group were Tianyan No.1, Qingtian No.2, and Tianyan No.3, while the three varieties with the lowest RFV and RFQ were Bayan No.6, Youmu No.1, and Titan; the three varieties with the highest GI were Qinghai 444, Tianyan No.1, and Tianyan No.3, while the three varieties with the lowest GI were Magnum, Bayan No.6, and Youmu No.1.

The three varieties with the highest RFV and RFQ in the IN group were Tianyan No.1, Qingtian No.2, and Everleaf 126, while the three varieties with the lowest RFV and RFQ were Monica, Youmu No.1, and Magnum; the three varieties with the highest GI were Qinghai 444, Titan, and Tianyan No.1, while the three varieties with the lowest GI were Hanma, Baler, and Youmu No.1.

3.4. In Vitro Digestibility of Silage from Different Oat Forage Varieties

By analysing the in vitro digestibility of silage from different oat forage varieties, the results are shown in

Table 4. Effect of lactic acid bacterial inoculants on in vitro fermentation parameters of silage from 16 oat forage varieties.

Variety	Treatment	DMD	pH	NH3-N (mg/dL)
Qingtian No.2	CK	66.53 ± 2.08 de	6.92 ± 0 c	13.35 ± 0.38 cd
	IN	71.92 ± 0.91 bc	6.99 ± 0.07 bcd	14.95 ± 0 bcde
Baiyan No.7	CK	66.87 ± 2.2 cd	6.99 ± 0.02 b	15.28 ± 0.38 b
	IN	67.13 ± 0.18 g	6.97 ± 0.01 d	15.8 ± 0.19 abcd
Baler	CK	71.04 ± 1.1 ab	6.82 ± 0.07 d	12.88 ± 0.21 de
	IN	73.88 ± 1.08 ab	6.88 ± 0.05 ef	13.83 ± 3.85 de
Haymaker	CK	72.61 ± 2.49 a	6.82 ± 0.05 d	11.1 ± 2.12 f
	IN	74.17 ± 0.35 ab	6.85 ± 0.02 f	12.95 ± 0 e
Hanma	CK	69.3 ± 1.06 bc	6.91 ± 0.02 c	12.2 ± 0 e
	IN	71.34 ± 0.97 cd	6.85 ± 0.05 f	13.55 ± 0 e
Bayan No.6	CK	64.33 ± 0.92 def	7.11 ± 0.05 a	13.38 ± 0.32 d
	IN	69.48 ± 1.1 def	7.08 ± 0.02 ab	14.31 ± 0.33 cde
Everleaf 126	CK	65.27 ± 1.31 def	6.82 ± 0.06 d	12.83 ± 0.12 de
	IN	69.26 ± 1.47 defg	6.86 ± 0.03 f	12.85 ± 0.06 e
Titan	CK	62.56 ± 0.18 f	6.92 ± 0.03 c	15.47 ± 0.13 b
	IN	67.36 ± 1.4 fg	7.06 ± 0.14 abc	16.55 ± 0.35 abc
Youmu No.1	CK	69.84 ± 1.93 b	6.87 ± 0.03 cd	13.14 ± 0.5 cde
	IN	72.62 ± 0.91 abc	6.83 ± 0.02 f	14.25 ± 1.92 cde
Sweet Oat	CK	65.2 ± 0.92 def	7 ± 0.02 b	13.42 ± 0.26 cd
	IN	72.5 ± 0.41 abc	6.88 ± 0.03 ef	15.17 ± 0 bcde
Magnum	CK	63.89 ± 0.41 ef	7.03 ± 0.04 b	15.8 ± 0.43 ab
	IN	68.34 ± 0.6 efg	6.98 ± 0.03 cd	17.19 ± 2.62 ab
Forage plus	CK	64.95 ± 1.13 def	6.93 ± 0.05 c	14.03 ± 0.8 c
	IN	70.82 ± 1.16 cd	6.96 ± 0.03 de	14.47 ± 0.45 cde
Qinghai 444	CK	62.52 ± 0.5 f	7.04 ± 0.05 b	16.51 ± 0.29 a
	IN	69.66 ± 1.25 de	7.08 ± 0.02 ab	17.63 ± 0.22 a
Tianyan No.3	CK	69.76 ± 1.68 b	6.84 ± 0.02 d	14.03 ± 0 c
	IN	72.63 ± 0.61 abc	6.85 ± 0.03 f	14.48 ± 0.12 cde
Monica	CK	64.73 ± 1.41 def	7 ± 0.03 b	16.26 ± 0.15 ab
	IN	67.97 ± 0.93 efg	7.09 ± 0.05 a	16.61 ± 0.57 abc
Tianyan No.1	CK	70.06 ± 2.06 ab	6.75 ± 0.02 e	13.79 ± 0.74 cd
	IN	74.77 ± 3.12 a	6.82 ± 0.02 f	15.41 ± 0.91 abcd
SEM		0.368	0.01	0.177
Inoculants		**	NS	**
Variety		**	**	**
Inoculants × Variety		**	**	NS

Note: The data in the table are mean ± standard error. Different lowercase letters in the same column indicate significant differences between the same indicators of different oat forage varieties. **: p < 0.01; NS: not significant; CK: control group (silage without lactic acid bacteria added); IN: bacteria added group (silage with lactic acid bacteria added); DMD: dry matter degradability.

. The LAB inoculant had a significant effect on DMD and NH₃-N (p < 0.01), and the difference was not significant for pH (p > 0.05); the variety had a highly significant effect on DMD, pH, and NH₃-N (p < 0.01). In addition, the two of the variety and LAB inoculant interaction pH and NH₃-N had highly significant effects (p < 0.01), and the difference on DMD was not significant (p > 0.05).

The DMD content of the CK group Haymaker was significantly higher than that of other varieties except Baler and Tianyan No.1 (p < 0.01), and that of Qinghai 444 was significantly lower than that of other varieties except Everleaf 126, Sweet oat, Forage plus, Monica, Bayan No.6, Magnum, and Titan (p < 0.01); the pH of Bayan No.6 was significantly higher than that of other varieties (p < 0.01), and pH of Tianyan No.1 was significantly lower than that of other varieties (p < 0.01). NH₃-N of Qinghai 444 was significantly higher than that of other varieties except Magnum and Monica (p < 0.01), and that of Haymaker was significantly lower than that of other varieties (p < 0.01).

The DMD content of Tianyan No.1 in the IN group was significantly higher than that of other varieties except for Haymaker, Baler, Tianyan No.3, Youmu No.1, and Sweet oat (p < 0.01), while the DMD content of Baiyan No.7 was significantly lower than that of other varieties except for Everleaf 126, Magnum, Monica, and Titan (p < 0.01). The pH of Monica was significantly higher than that of other varieties except for Bayan No.6, Qinghai 444, and Titan (p < 0.01). The pH of Tianyan No.1 was significantly lower than that of other varieties except for Baler, Sweet oat, Everleaf 126, Hanma, Tianyan No.3, Haymaker, and Youmu No.1 (p < 0.01), and NH₃-N of Qinghai 444 was significantly higher than that of other varieties except for Haymaker, Monica, Titan, Baiyan No.7, and Tianyan No.1 (p < 0.01). The NH₃-N of Everleaf 126 was significantly lower than that of other varieties except for Sweet oat, Qingtian No.2, Tianyan No.3, Forage, Bayan No.6, Youmu No.1, Baler, Hanma, and Haymaker (p < 0.01).

From Figure 1 and Figure 2, it can be seen that the gas production of each variety increased with the increase in fermentation time. In the CK group, the three varieties with the highest total gas production were Haymaker, Tianyan No.1, and Youmu No.1, and the three varieties with the lowest total gas production were Qingtian No.2, Monica, and Magnum. In the IN group, the three varieties with the highest total gas production were Haymaker, Tianyan No.3, and Baiyan No.7, and the three varieties with the lowest total gas production were Titan, Qingtian No.2, and Magnum. All varieties had the highest gas production rate between 2 h and 12 h, and the trend of gas production increased slowly after 48 h.

3.5. Comprehensive Evaluation

A principal component analysis was conducted on the standardised nutritional quality data (excluding silage pH and rumen pH) after normalisation, and the standardised data are shown in

Table 5. Standardised data of oat forage silage indicators.

Variety	Treatment	CP	Ash	WSC	NDF	ADF	RFV	RFQ	GI	DMD	NH3-N	GP	LA	AA
CK	Qingtian No.2	−1.063	−0.629	2.776	−1.110	−0.779	1.099	1.099	0.236	−0.671	−0.731	−1.587	−1.135	0.490
	Baiyan No.7	−0.680	−0.482	0.924	0.414	0.807	−0.491	−0.491	−0.748	−0.572	0.513	−0.206	−0.730	0.313
	Baler	−0.818	1.358	−0.288	0.690	0.240	−0.613	−0.613	−1.039	0.631	−1.034	−0.141	−1.509	−0.220
	Haymaker	−0.388	−0.555	0.735	0.617	0.299	−0.587	−0.586	−0.737	1.084	−2.181	1.169	−1.869	0.366
	Hanma	−1.002	−0.169	−0.019	0.164	−0.258	−0.084	−0.083	−0.748	0.129	−1.472	−0.076	−1.944	1.661
	Bayan No.6	0.518	−0.243	−1.070	2.196	2.068	−2.084	−2.084	−1.473	−1.306	−0.711	−1.380	0.936	−0.379
	Everleaf 126	−1.048	−0.335	−1.490	0.623	0.713	−0.737	−0.737	−1.254	−1.034	−1.066	0.215	−1.244	2.087
	Titan	1.624	−0.978	−0.660	0.549	1.866	−1.226	−1.227	0.006	−1.816	0.636	−0.188	0.292	0.916
	Youmu No.1	−0.695	0.438	−0.326	1.081	1.472	−1.281	−1.281	−1.435	0.285	−0.866	0.430	−0.705	−0.344
	Sweet Oat	0.580	−0.040	−0.788	0.264	0.895	−0.570	−0.571	−0.044	−1.055	−0.686	−0.461	−1.634	1.430
	Magnum	1.102	−0.702	0.429	1.608	0.038	−1.100	−1.100	−1.996	−1.433	0.849	−2.129	−0.443	0.419
	Forage plus	−0.818	−0.629	−1.023	0.231	−0.212	−0.146	−0.146	−0.676	−1.127	−0.292	−0.903	0.543	1.004
	Qinghai 444	1.701	2.719	0.452	−0.331	−0.004	0.178	0.178	1.484	−1.828	1.306	−1.338	−1.244	−0.557
	Tianyan No.3	−0.081	−0.905	−0.019	−0.851	−0.414	0.737	0.737	0.731	0.262	−0.292	−0.217	0.331	1.306
	Monica	1.133	0.714	−0.476	0.337	1.719	−0.916	−0.916	0.011	−1.190	1.145	−1.697	−0.327	−0.326
	Tianyan No.1	−1.217	−1.438	−0.222	−2.160	−0.695	2.019	2.018	1.012	0.348	−0.447	0.517	1.208	1.749
IN	Qingtian No.2	−0.925	−0.463	2.899	−1.310	−1.731	1.651	1.651	0.797	0.885	0.301	−0.445	0.942	−0.628
	Baiyan No.7	−0.541	−0.206	1.079	0.117	−0.981	0.225	0.226	−0.198	−0.497	0.849	1.195	0.180	−0.823
	Baler	−0.618	1.597	−0.208	0.599	−1.138	−0.067	−0.067	−0.594	1.451	−0.421	0.616	0.016	−1.497
	Haymaker	−0.280	−0.426	0.910	0.155	−0.340	−0.046	−0.046	−0.214	1.535	−0.988	1.968	0.158	−1.497
	Hanma	−0.833	−0.040	0.019	−0.154	−0.779	0.293	0.293	−0.319	0.718	−0.602	0.388	0.551	−0.947
	Bayan No.6	0.626	0.070	−1.056	0.493	−0.385	−0.227	−0.227	0.231	0.181	−0.112	0.407	1.699	−0.592
	Everleaf 126	−0.910	−0.040	−1.433	−1.057	−1.232	1.233	1.232	0.440	0.117	−1.053	0.894	0.521	0.543
	Titan	1.685	−0.905	−0.519	−0.360	−0.884	0.537	0.538	1.814	−0.431	1.332	−0.230	1.372	−0.929
	Youmu No.1	−0.480	0.769	−0.118	0.187	1.237	−0.643	−0.643	−0.770	1.087	−0.150	0.794	0.860	−1.106
	Sweet Oat	0.749	0.328	−0.481	−0.022	0.569	−0.255	−0.256	0.374	1.053	0.443	0.026	−0.291	−0.698
	Magnum	1.271	−0.592	0.452	1.502	−0.454	−0.876	−0.876	−0.088	−0.148	1.745	−1.343	0.155	−0.663
	Forage plus	−0.618	−0.445	−0.901	−0.060	−0.805	0.282	0.282	−0.198	0.568	−0.009	−0.192	0.636	0.171
	Qinghai 444	1.824	2.792	0.504	−0.757	−0.460	0.681	0.681	2.166	0.233	2.028	0.205	−0.008	−1.248
	Tianyan No.3	0.027	−0.684	0.287	−1.237	−0.799	1.209	1.209	1.287	1.090	−0.002	1.572	1.236	0.064
	Monica	1.255	1.026	−0.302	−0.033	1.345	−0.535	−0.535	0.484	−0.255	1.371	0.983	0.475	−1.071
	Tianyan No.1	−1.079	−0.905	−0.066	−2.384	−0.919	2.341	2.341	1.457	1.708	0.597	1.153	0.970	1.004

Note: CK: control group (silage without lactic acid bacteria added), IN: bacteria added group (silage with lactic acid bacteria added); CP: crude protein; WSC: water-soluble carbohydrate; NDF: neutral detergent fibre; ADF: acid detergent fibre; Ash: crude ash; RFV: relative feeding value; RFQ: relative forage quality; GI: forage grading index; DMD: dry matter degradability; GP: gas production; LA: lactic acid; AA: acetic acid.

. The principal components were extracted according to the principle of eigenvalue > 1. The results are shown in

Table 6. Loading factor, eigenvalue, and contribution rate of principal component.

Index	Main Ingredient
	1	2	3	4
CP	−0.398	0.818	−0.109	0.13
Ash	−0.202	0.492	0.424	−0.363
WSC	0.353	0.113	−0.005	−0.752
NDF	−0.916	−0.053	0.22	−0.011
ADF	−0.8	−0.001	0.005	0.203
RFV	0.975	0.024	−0.182	−0.054
RFQ	0.975	0.024	−0.182	−0.054
GI	0.686	0.596	−0.164	0.084
DMD	0.604	−0.227	0.665	0.019
NH3-N	0.012	0.902	−0.178	0.048
GP	0.486	−0.239	0.613	0.307
LA	0.419	0.335	0.097	0.664
AA	0.03	−0.589	−0.721	0.134
Eigenvalue	4.951	2.664	1.714	1.323
Contribution rate (%)	38.083	20.491	13.182	10.173
Cumulative contribution rate (%)	38.083	58.574	71.756	81.929

Note: CP: crude protein; WSC: water-soluble carbohydrate; NDF: neutral detergent fibre; ADF: acid detergent fibre; Ash: crude ash; RFV: relative feeding value; RFQ: relative forage quality; GI: forage grading index; DMD: dry matter degradability; GP: gas production; LA: lactic acid; AA: acetic acid.

: principal component 1, with an eigenvalue of 4.951 and a contribution rate of 38.083%; principal component 2, with an eigenvalue of 2.664 and a contribution rate of 20.491%; principal component 3, with an eigenvalue of 1.714 and a contribution rate of 13.182%; and principal component 4, with an eigenvalue of 1.323 and a contribution rate of 10.173%.The cumulative contribution rate of the four principal components was 81.929%, containing 81.929% of the information of the measured indicators.

As shown in Table 6, the 1st principal component includes RFV, RFQ, NDF, and ADF, which are related to the fibre content; the 2nd principal component includes NH₃-N, CP, AA, and GP, which are related to the protein content; the 3rd principal component is mainly Ash; and the 4th principal component is mainly LA and WSC.

The formula for the score value of each principal component is as follows:

$$\begin{gathered} \mathrm{F}1=-0.398\mathrm{Y}1-0.202\mathrm{Y}2+0.353\mathrm{Y}3-0.916\mathrm{Y}4-0.8\mathrm{Y}5+0.975\mathrm{Y}6+0.875\mathrm{Y}7+0.686\mathrm{Y}8+0.604\mathrm{Y}9 \\ +0.012\mathrm{Y}10+0.486\mathrm{Y}11+0.419\mathrm{Y}12+0.03\mathrm{Y}13 \end{gathered}$$

$$\begin{gathered} \mathrm{F}2=0.818\mathrm{Y}1+0.492\mathrm{Y}2+0.113\mathrm{Y}3-0.053\mathrm{Y}4-0.001\mathrm{Y}5+0.024\mathrm{Y}6+0.024\mathrm{Y}7+0.596\mathrm{Y}8-0.227\mathrm{Y}9 \\ +0.902\mathrm{Y}10-0.239\mathrm{Y}11+0.335\mathrm{Y}12-0.589\mathrm{Y}14 \end{gathered}$$

$$\begin{gathered} \mathrm{F}3=-0.109\mathrm{Y}1+0.424\mathrm{Y}2-0.005\mathrm{Y}3+0.22\mathrm{Y}4+0.005\mathrm{Y}5-0.182\mathrm{Y}6-0.182\mathrm{Y}7-0.164\mathrm{Y}8+0.665\mathrm{Y}9-0.178\mathrm{Y}10 \\ +0.613\mathrm{Y}11+0.097\mathrm{Y}12-0.721\mathrm{Y}15 \end{gathered}$$

$$\begin{gathered} \mathrm{F}4=0.13\mathrm{Y}1-0.363\mathrm{Y}2-0.752\mathrm{Y}3-0.011\mathrm{Y}4+0.203\mathrm{Y}5-0.054\mathrm{Y}6-0.054\mathrm{Y}7+0.084\mathrm{Y}8+0.019\mathrm{Y}9+0.048\mathrm{Y}10 \\ +0.307\mathrm{Y}11+0.664\mathrm{Y}12+0.134\mathrm{Y}16 \end{gathered}$$

where F1, F2, F3, and F4 represent the 1st, 2nd, 3^rd, and 4th principal component scores, respectively; Y1 to Y13 represent the standardised values of CP to AA in that order.

The results are shown in

Table 7. Principal component scores of silage of 16 oat forage varieties.

Variety	Treatment	F1	F2	F3	F4
Qingtian No.2	CK	3.72	−1.41	−2.6	−3.47
Baiyan No.7	CK	−2.49	−0.97	−0.63	−0.95
Baler	CK	−3.06	−2.14	1.74	−1.47
Haymaker	CK	−1.44	−4.34	1.66	−1.26
Hanma	CK	−0.93	−4.33	−0.86	−1.32
Bayan No.6	CK	−10.15	−0.39	0.12	1.55
Everleaf 126	CK	−4.4	−4.43	−1.4	0.67
Titan	CK	−5.97	1.28	−2.08	1.82
Youmu No.1	CK	−5.38	−2.35	2	−0.12
Sweet Oat	CK	−4.05	−1.36	−1.85	−0.17
Magnum	CK	−7.12	0.48	−2.09	−0.86
Forage plus	CK	−1.55	−1.72	−1.88	0.99
Qinghai 444	CK	−1.69	5.51	−1.4	−2.28
Tianyan No.3	CK	3.41	−0.93	−1.77	0.56
Monica	CK	−5.81	2.96	−1.24	−0.06
Tianyan No.1	CK	8.67	−2.14	−2.37	1.46
Qingtian No.2	IN	8.37	0.83	−0.33	−2.14
Baiyan No.7	IN	1.94	0.6	0.8	−0.64
Beile2	IN	0.81	−0.1	3.49	−0.78
Haymaker	IN	2.31	−1.25	3.43	−0.16
Hanma	IN	2.25	−0.92	1.54	0.02
Bayan No.6	IN	−0.04	1.21	1.1	1.98
Everleaf 126	IN	5.13	−1.88	−0.2	1.28
Titan	IN	2.8	4.36	−1	1.63
Youmu No.1	IN	−1.52	−0.16	2.9	0.68
Sweet Oat	IN	−0.67	1.4	1.2	0.28
Magnum	IN	−3.6	3	−0.47	−0.3
Forage plus	IN	1.62	−0.9	0	0.95
Qinghai 444	IN	2.92	6.75	1.03	−1.14
Tianyan No.3	IN	7.06	0.36	0.54	1.2
Monica	IN	−2.01	3.61	1.42	0.92
Tianyan No.1	IN	10.87	−0.62	−0.78	1.14

Note: F1, F2, F3, and F4 represent the 1st, 2nd, 3rd, and 4th principal component scores. CK: control group (silage without lactic acid bacteria added); IN: bacteria added group (silage with lactic acid bacteria added).

.

The membership function values (μ) for the principal component scores in Table 7 were calculated, and the weights (w) for each principal component were determined based on their contribution rates in Table 6. The integrated score D was calculated and ranked according to w and μ. The higher the value D represented, the higher the nutritive value of the variety was, and the results are shown in

Table 8. Membership function values, comprehensive scores, and rankings of principal component scores of oat forage silage.

Variety	Treatment	μ1	μ2	μ3	μ4	D Value	Sequence
Tianyan No.1	IN	0.660	0.270	0.000	0.000	0.703	1
Qinghai 444	IN	0.364	0.309	0.323	0.462	0.688	2
Tianyan No.3	IN	0.337	0.205	0.713	0.367	0.677	3
Titan	IN	0.414	0.008	0.700	0.406	0.641	4
Qingtian No.2	IN	0.439	0.009	0.286	0.394	0.617	5
Tianyan No.1	CK	0.000	0.361	0.447	0.921	0.586	6
Haymaker	IN	0.274	0.000	0.197	0.760	0.581	7
Bayan No.6	IN	0.199	0.511	0.085	0.971	0.572	8
Everleaf 126	IN	0.227	0.186	0.755	0.615	0.567	9
Monica	IN	0.290	0.275	0.123	0.606	0.566	10
Baler	IN	0.144	0.439	0.084	0.479	0.561	11
Hanma	IN	0.409	0.242	0.118	0.818	0.542	12
Baiyan No.7	IN	0.402	0.889	0.197	0.218	0.534	13
Youmu No.1	IN	0.645	0.313	0.136	0.739	0.526	14
Sweet Oat	IN	0.206	0.661	0.223	0.626	0.526	15
Forage plus	IN	0.895	0.205	0.038	0.905	0.509	16
Tianyan No.3	CK	0.881	0.470	0.373	0.244	0.492	17
Qinghai 444	CK	0.575	0.450	0.558	0.519	0.468	18
Magnum	IN	0.521	0.387	1.000	0.494	0.440	19
Monica	CK	0.593	0.284	0.990	0.607	0.375	20
Qingtian No.2	CK	0.590	0.314	0.680	0.640	0.374	21
Forage plus	CK	0.481	0.504	0.608	1.000	0.371	22
Baler	CK	0.727	0.228	0.394	0.872	0.368	23
Haymaker	CK	0.616	0.786	0.263	0.936	0.358	24
Baiyan No.7	CK	0.411	0.382	0.903	0.761	0.356	25
Titan	CK	0.451	0.521	0.624	0.688	0.354	26
Youmu No.1	CK	0.312	0.665	0.350	0.582	0.350	27
Hanma	CK	0.560	0.316	0.427	0.811	0.301	28
Sweet Oat	CK	0.622	1.000	0.596	0.428	0.299	29
Bayan No.6	CK	0.819	0.428	0.516	0.857	0.277	30
Everleaf 126	CK	0.387	0.719	0.660	0.806	0.253	31
Magnum	CK	1.000	0.341	0.299	0.846	0.250	32
Weight		0.566	0.285	0.149

Note: μ1, μ2, μ3, and μ4 represent the membership function values. CK: control group (silage without lactic acid bacteria added); IN: bacteria added group (silage with lactic acid bacteria added).

.

From the rankings in Table 7, it can be seen that the three varieties with the best silage quality in the IN group were Tianyan No.1, Qinghai 444, and Tianyan No.3, and the three varieties with the lowest silage quality were Sweet oat, Forage plus, and Magnum; the three varieties with the best silage quality in the CK group were Tianyan No.1, Tianyan No.3, and Qinghai 444, and the three varieties with the lowest silage quality were Bayan No.6, Everleaf 126, and Magnum.

4. Discussion

4.1. Effect of Lactic Acid Bacterial Inoculants on Silage Nutrient Composition of Different Oat Forage Varieties

The nutrient content of silage are crucial indicators for evaluating the nutritional value of oat forage silage. CP content is one of the key parameters for assessing forage nutritional quality, while Ash content reflects the inorganic matter content in silage. NDF content and ADF content influence both the palatability and digestibility of silage. Generally, higher contents of CP, Ash, and WSC, along with lower levels of NDF and ADF, indicate superior silage quality. After ensiling fermentation, the nutrient content of oat forage silage varied among different varieties. The CP content ranged from 4.3% to 6.2% DM in the CK treatment and from 4.39% to 6.28% DM in the IN treatment. Compared with the CK treatment, the CP content of each oat forage variety in the IN treatment showed varying degrees of increase, indicating that the addition of LAB inoculant reduced the amount of nitrogen loss during the ensiling process. The Ash content ranged from 4.2% to 6.46% DM in the CK group and from 4.49% to 6.5% DM in the IN group. The ash content in the IN group was higher than that in the CK group, which was consistent with the findings of Wei et al. [20]. The WSC content ranged from 3.1% to 12.15% DM in the CK treatment and from 3.22% to 12.41% DM in the IN treatment. Among the 16 oat forage silage varieties tested, the WSC content was all above 3%, which ensured good preservation of the silage [21]. The WSC content in the IN group was higher than that in the CK group, which might be attributed to the reduction in pH by LAB during the fermentation process, inhibiting the growth of harmful microorganisms and thereby reducing the consumption of WSC. However, during the fermentation process, WSC will continue to be converted into LA, and there was a secondary fermentation conversion, so the content increase was not obvious [22]. The NDF content in the CK treatment ranged from 44.31% to 59.12% DM, and the ADF content ranged from 22.29% ~31.03% DM. The NDF content ranged from 43.55% to 56.76% DM, and the ADF content ranged from 19.37% to 28.81% DM in the IN treatment. The IN treatment showed a decreasing trend of NDF and ADF contents compared to the CK treatment, which was consistent with the findings of Xu et al. [23].

4.2. Effect of Lactic Acid Bacterial Inoculants on Silage Quality of Different Oat Forage Varieties

pH is one of the critical indicators for evaluating the quality of silage. A lower pH can ensure good preservation of silage, and the pH is directly related to the number of LAB in silage [24]. The fermentation of LAB can rapidly reduce pH, inhibit protein hydrolysis by other harmful microorganisms in a high acid concentration condition, and improve the quality and palatability of the silage [25]. In this study, the silage pH content ranged from 3.6 to 3.95 in the CK treatment and from 3.68 to 3.83 in the IN treatment, both of which were acidic and favourable for silage fermentation. Shao et al. [26] concluded that a pH lower than 4.2 meets the criteria for high-quality silage. Compared to the CK treatment, the pH of Qingtian No.2, Baiyan No.7, Baler, Haymaker, Hanma, Bayan No.6, Titan, Youmu No.1, and Sweet oat in the IN treatment showed varying degrees of reduction, indicating that inoculation with LAB significantly reduced the pH of oat forage silage.

Organic acid is one of the key indicators for evaluating the fermentation quality of silage oat forages. A higher LA content indicates better inhibition of the growth and reproduction of harmful microorganisms and effectively prevents the spoilage of silage oat forages [22]. The content of AA in silage mainly reflects the aerobic stability and preservation performance of silage [27]. The PA and BA content should be as little as possible in silage oat forages, especially BA produced by harmful microorganisms such as Clostridium, which not only leads to the decomposition of proteins in silage but also produces a bad odour, which seriously affects the fermentation quality of silage oat forages [28]. In this study, the WSC content of the raw materials of various oat forage varieties ranged from 3.49% to 12.61% DM. The decrease in the WSC content after ensiling indicates that a significant amount of WSC was consumed as a fermentation substrate, indicating successful silage [29]. LAB under anaerobic conditions utilised WSC as the main fermentation substrate to produce large amounts of LA, which contributed to the decrease in pH. The lowest pH and higher LA content were obtained after the addition of LAB inoculants, which resulted in the improvement of the fermentation quality of silage. The range of the LA content in the CK group ranged from 28.97 mg/g to 48.21 mg/g and that in the IN group ranged from 38.06 mg/g to 51.53 mg/g. The LA in all the oat forage varieties under the IN treatment was higher than that of the CK treatment, which indicated that the addition of LAB favoured fermentation and promoted LA production. The range of the AA content was 1.78–3.27 mg/g in the CK group and 1.25–2.66 mg/g in the IN group. BA and PA was not detected in all the samples, which indicated that the fermentation was good, likely due to the reduction in pH and the accumulation of acidic substances, which inhibited harmful microorganisms [30]. This is in agreement with the study of Li et al. [31]. Therefore, while the nutritional composition of silage varies among different oat forage varieties, the addition of LAB inoculants enables the preparation of high-quality silage across all varieties.

4.3. Effect of Lactic Acid Bacterial Inoculants on Feed Quality Rating Index of Different Oat Forage Varieties

In the experiment, the RFV ranged from 101.87 to 149.77, RFQ ranged from 131.59 to 224.99, and GI ranged from 6.32 MJ/d to 12.65 MJ/d in the CK treatment, while the RFV ranged from 115.97 to 153.53, RFQ ranged from 159.09 to 232.33, and GI ranged from 8.55 MJ/d to 13.89 MJ/d in the IN treatment. Compared to the CK treatment RFV, RFQ, and GI were significantly higher in the IN treatment (p < 0.01). The overall trend of RFQ was the same as that of RFV. The RFQ range of the 16 oat forage varieties after ensiling with LAB inoculants was overall higher than that reported in the central and southern regions of Heilongjiang (128.28~208.78) [32], the northwestern region of Hebei (111.00–181.50) [33], and the Shigatse region of Tibet (118.03–174.76) [34], indicating that the quality of oat forage silage was significantly improved with the addition of LAB. In this experiment, the GI of 16 oat forage varieties ranged from 6.34 MJ/d to 11.91 MJ/d. Qinghai 444 had the highest GI, followed by Tianyan No.1 and Tianyan No.3, while Everleaf 126 had the lowest GI, followed by Bayan No.6. According to the GI classification standard [35], eight oat forage varieties belonged to Grade 1 forage, namely Qinghai 444, Tianyan No.1, Tianyan No.3, Sweet oat, Qingtian No.2, Monica, Haymaker, and Magnum, and the remaining eight oat forage varieties belonged to Grade 2 forage. Except for the 4 oat forage varieties in the CK treatment, namely Everleaf 126, Youmu No.1, Bayan No.6, and Magnum, which belonged to Grade 2 forage, all the forages in the IN treatment and the 12 varieties in the CK treatment, namely Qingtian No.2, Baiyan No.7, Baler, Haymaker, Hanma, Titan, Sweet oat, Forage plus, Qinghai 444, Tianyan No.3, Monica, and Tianyan No.1, belonged to Grade 1 forage. Considering the comprehensive performance of RFV, RFQ, and GI, Tianyan No.1 exhibited outstanding results, indicating its suitability as a silage forage variety for local silage processing. Furthermore, the addition of LAB inoculants significantly improved the silage quality of oat forage.

4.4. Effect of Lactic Acid Bacterial Inoculants on In Vitro Digestibility of Silage from Different Oat Forage Varieties

LAB inoculants improved silage quality and digestibility by rapidly lowering pH through LA production, optimising the acid profile, and ultimately increasing DMD [36]. In the CK treatment, the rumen pH ranged from 6.75 to 7, and the rumen NH₃-N concentration ranged from 11.1 mg/dL to 16.51 mg/dL. In the IN treatment, the rumen pH ranged from 6.82 to 7.09, and the rumen NH₃-N concentration ranged from 12.85 mg/dL to 17.63 mg/dL. The rumen pH and NH₃-N concentration in all groups of this study were within the normal range, indicating that LAB and silage of different oat forage varieties had no adverse effects on the rumen internal environment, which is consistent with the study of Zhang et al. [37].

In this experiment, DMD was significantly improved in the IN treatment compared to the CK treatment (p < 0.01), and the DMD of oat forage silage with LAB inoculants basically reached more than 67.13%, which was similar to the results reported by Liu et al. [38]. Among the CK and IN treatments, the three varieties with the highest dry matter degradation rates were Haymaker, Baler, and Tianyan No.1, which may be associated with the lower average contents of NDF and ADF.

Gas production volume and gas production rate can reflect the degradation rate and rumen microbial activity [39]. In this experiment, there were significant differences in the gas production characteristics of different varieties of oat forages after in vitro fermentation. The study found that the gas production rates of different oat forage varieties varied significantly between 2 h and 8 h. The gas production rates of Tianyan No.1 and Haymaker increased faster than those of other varieties. The gas production rates of all varieties increased again from 24 h to 36 h, and the rest of the time period was relatively slow. The total gas production of all IN treatments in this experiment was higher than that of the CK treatment, suggesting that the addition of LAB inoculants may increase the activity of rumen microorganisms and promote the degradation of oat forage silage in the rumen [40,41,42]. In the CK treatment, the three varieties with the highest cumulative gas production at 72 h were Haymaker, Tianyan No.1, and Youmu No.1, while the three varieties with the highest cumulative gas production at 72 h in the IN treatment were Haymaker, Tianyan No.3, and Baiyan No.7.

4.5. Integrated Evaluation of Different Oat Forage Varieties for Silage

In this experiment, the 13 indicators determined by the two treatments were combined into four independent factors by principal component analysis and membership function methods. These factors reflected 81.929% of the original variable information. Based on the principal component scores, the comprehensive scores of different treatment qualities were calculated after processing with the membership function. The results of the study showed that the IN treatment was superior to the CK treatment. Specifically, the Tianyan No.1 variety in the IN group obtained the highest comprehensive score of 0.703, followed by Qinghai 444 and Tianyan No.3 in the IN group, with comprehensive scores of 0.688 and 0.677, respectively. These data indicate that the addition of LAB inoculants significantly enhances the quality of silage.

5. Conclusions

Our results confirmed that silage from different oat forage varieties showed variations in nutrient composition, fermentation quality, and in vitro digestibility. Ensiling improved the nutritional value of oat forage. Adding LAB inoculants enhanced the nutritional quality, fermentation, and digestibility of silage. Among the 16 oat forage varieties tested, Tianyan No.1, Qinghai 444, and Tianyan No.3 not only exhibited high comprehensive nutritional value but also demonstrated excellent potential feeding value, making them suitable varieties for cultivation and processing in the Qinghai–Tibet Plateau.